TOTTEN ATTUT IT US 20180268942A1BEI KITUMIA ULE TUTTI ( 19) United States ( 12) Patent Application Publication ( 10 ) Pub. No. : US 2018 /0268942 A1 KAMALI -ZARE et al. (43 ) Pub . Date : Sep . 20 , 2018 (54 ) METHODS AND SYSTEMS FOR Publication Classification IDENTIFYING BRAIN DISORDERS (51 ) Int. CI. G16H 50/ 50 (2006 . 01 ) (71 ) Applicant: DARMIYAN , INC ., Emeryville , CA A61B 5 / 00 ( 2006 .01 ) (US ) A61B 5 / 055 ( 2006 .01 ) G16H 30/ 40 ( 2006 . 01 ) (72 ) Inventors : Padideh KAMALI -ZARE , Emeryville , G06T 17 / 10 ( 2006 . 01 ) CA (US ) ; Kaveh VEJDANI, G16H 50/ 70 ( 2006 .01 ) Emeryville , CA (US ) ; Thomas G06T 7 / 00 (2006 .01 ) LIEBMANN , Emeryville , CA (US ); (52 ) U . S . CI. Hesaam ESFANDYARPOUR , CPC ...... G16H 50 /50 ( 2018 .01 ) ; A61B 5 / 0042 Emeryville , CA (US ) ( 2013 .01 ); A61B 5 / 055 ( 2013 .01 ); A61B 5 / 7278 ( 2013 .01 ) ; A61B 5 / 4088 ( 2013 .01 ) ; A61B 5 / 4082 ( 2013 . 01) ; G06T 2207 / 10088 ( 21 ) Appl. No. : 15 / 987, 794 ( 2013 .01 ) ; G16H 30 / 40 (2018 .01 ) ; G06T | 17/ 10 ( 2013. 01 ) ; G16H 50 / 70 ( 2018. 01 ) ; GO6T 770016 ( 2013 .01 ); G06T 2207 / 30016 ( 22 ) Filed : May 23 , 2018 ( 2013 .01 ) ; A61B 5 /4094 (2013 .01 ) (57 ) ABSTRACT Related U . S . Application Data Methods and systems for determining whether brain tissue is indicative of a disorder , such as a neurodegenerative disor (63 ) Continuation of application No . PCT /US2017 / der, are provided . The methods and systems generally utilize 064745 , filed on Dec . 5 , 2017 . data processing techniques to assess a level of congruence (60 ) Provisional application No . 62/ 481 ,839 , filed on Apr. between measured parameters obtained from magnetic reso 5 , 2017 , provisional application No. 62 / 430 , 351, filed nance imaging (MRI ) data and simulated parameters on Dec . 6 , 2016 . obtained from computational modeling of brain tissues.

Healthy ?????????? Patent Application Publication Sep . 20 , 2018 Sheet 1 of 20 US 2018 / 0268942 A1

100 Obtain magnetic resonance imaging (MRI ) data comprising one or more measured MRI parameters . 110

Use one or more computer processors to process the one or more measured MRI parameters with one or more simulated MRI 120 parameters generated from one or more microstructuralmodels .

Select a diagnostic model from the one ormore nicrostructuralmodels based on a threshold congruence between the one or more measured 130 MRI parameters and the one or more simulated MRI parameters

Use the diagnostic model to determine the disorder state of brain tissue in a brain of a subject. 140

FIG . 1 Patent Application Publication Sep . 20 , 2018 Sheet 2 of 20 US 2018 / 0268942 A1

200 -

Create a firstmicrostructural model corresponding to a brain state that is not associated with a disorder . 210

iteratively subject the first nicrostructuralmodel to a perturbation . 220

FIG . 2 Patent Application Publication Sep . 20 , 2018 Sheet 3 of 20 US 2018 / 0268942 A1

310

320

FIG . 3 Patent Application Publication Sep . 20 , 2018 Sheet 4 of 20 US 2018 / 0268942 A1

430

Network

415 435 401 440 405 11420

11425

* 1932FIG . 4 Patent Application Publication Sep . 20, 2018 Sheet 5 of 20 US 2018/ 0268942 A1

520

5?

FIG . 5A FIG. 5B

Healthy Degenerating

? FIG. 50: Patent Application Publication Sep . 20 , 2018 Sheet 6 of 20 US 2018 / 0268942 A1

p

DDDGray PND 620 6204 1630 640

610 FIG . 6 Patent Application Publication Sep . 20 , 2018 Sheet 7 of 20 US 2018 / 0268942 A1

MRI

AND

FIG . 7 Patent Application Publication Sep . 20 , 2018 Sheet 8 of 20 US 2018 / 0268942 A1

Grayscale MRI VYVYN PND Maps QND Maps 0000000000 00000000000 00000000OOOO Aged/AsymptomaticYoung GALOG OBULI: DODATDA * DOPODDENDO OOOOOOOO BORDO Aged/Symptomatic OOOOOOOOO

-

-

- 203 - 3629092 FIG . 8 3303 Patent Application Publication Sep . 20 , 2018 Sheet 9 of 20 US 2018 /0268942 A1

CROU 01020304050 01020304050 man gemeentede en men po5.1 %5.1 0.5 1.56 54.0 0.1A 0.5

wwwwwwww w www

10203040500 1020304050 OROU 01020304050

3.00 2.4 2. 3.5pm 3.0 2.0 2.81 612. 2.51

END 10203040500 FIG.9

1020304050 OND 01020304050

01020304050

etenerteret

r 010203040500 -MCI -AD PND prostorantementeentmmNormal ior inter

GROUPED FEMALE MALE Patent Application Publication _ Sep . 20, 2018 Sheet 10 Of 20 _ US 2018/ 026894241

1-?

-

- 5CROU -

/0? n ???15 ;.5 0,5 . oli?1 ;

GROU ????????" ??? _ 0203040500_44050°0f060°050 6 ? ? ? ? ? ? ; ?? ?? ? ??? ? dbi?

FIG.9{Cont) END 02030405010_50°050°°0 0_102030405050°0°°0'

: S = N – ??? 4 ?

QND

1- Male ???- PND -efmiae "020304050

4 ? NORMAL Patent Application Publication Sep . 20 , 2018 Sheet 11 of 20 US 2018 /0268942 A1

Abnormal,male Normal,male 12-16

NaNam 8-12 WAbnormal,femaleD AVNormal,femaleZ

8-4 Yearsbeforediagnosis FIG.10 WAbnormal,all WANormal,all

Om

% 3 3 0 : ?I _ number Scan Patent Application Publication Sep . 20 , 2018 Sheet 12 of 20 US 2018 / 0268942 A1

subwohnt n12*10FIG.11 12-16 12-16 12-16 N

w

w 04-812 O4-812 w Normal,Al Yearsbeforefinalscan Normal,Female Yearsbeforefinalscan Normal,Male 0-4812Yearsbeforefinalscan BAION wwwwwwwwwwwwwwwwwww

0 0 0

lum mtLLLL ...... i a è ?? ?? ? 10 * value Measured }{ { * value ??asured 10 * value Measured MWPNDWZQNDDENDVgRouAWRou leukunimente www websites

12-16 12-16 12-16

Om4-812 Abnormal,All 0-4812Yearsbeforediagnosis Abnormal,Female Yearsbeforediagnosis Abnormal,Male 0-4812Yearsbeforediagnosis

0

- / / LLLLLL| |LL | |LALU | | ?? 10 * value Measured 10 * value Measured 10 * value Measured Patent Application Publication Sep . 20 , 2018 Sheet 13 of 20 US 2018 / 0268942 A1

gm15 VIIDeterminedNormal DeterminedAbnormal

6-9

inn 3-6 Yearsbeforediagnosis FIG.12

0-3

Scans Brain of Number Patent Application Publication Sep . 20 , 2018 Sheet 14 of 20 US 2018 / 0268942 A1

DOO LOOD OODOO

FIG . 13 Patent Application Publication Sep . 20 , 2018 Sheet 15 of 20 US 2018 / 0268942 A1

Region Optimization

.

. PredictionAccuracy(%) rrrrrrrrrrrrr. 888990 'rovision . Provinceof .. . Basalganglia WholeCerebellumbrain PosteriorThalamus cingulate Precuneus HippocampusEntorhinal cortex Parietal OccipitalPrefrontallobe lobePremotor cortex Precentral cortexPostcentral gyrus TemporalParacentralgyrus Anteriorlobe Olfactory lobule - mid bulbcingulum Brain Region

FIG . 14 Patent Application Publication Sep . 20 , 2018 Sheet 16 of 20 US 2018 / 0268942 A1

Whole Brain

© Normal - AD Storago 0 .83* PNDvalue *** ** * PNDvalue ar0.0 pe the . Fiestas " ...... 50 60 70 80 90 100 50 60 70 80 90 100 Age Age 100 Posterior Cingulate g potemse OO de * {??? og PNDvalue . 8

@ - , - , - , 60 70 80 90 100 50 60 70 80 90 100 Age Age Precuneus S

tar 8 2025 35 gmaon@ Com {??3c @ PNDvalue *CO 08 0 08. #od T***

40 SO 60 70 80 90 100 50 60 70 80 90 100 Age Age

FIG . 15 Patent Application Publication Sep . 20 , 2018 Sheet 17 of 20 US 2018 / 0268942 A1

Entorhinal Cortex Basal Ganglia Normal AD 86 8 800 * * Top20 anjenond a PNDvalue *** 7 3 39 . One * * .

A L ...... -. . .- . - .- .- ...... - - - .- - 50 60 70 80 90 100 50 60 70 80 90 100 Ace Age Parietal lobe Occipital Lobe

: o o

o PNDvalue PNOvalue Co .

------50 60 70 80 90 100 50 60 70 80 90 100 ?????? Age Premotor Cortex go

.. ** 8 Co lescone . 100ncanto , 000 . . gh?khi PNDvalue * 0 DE 0 To ; 000coo " OS SO 60 70 80 90 100 50 60 70 80 90 100 Age Age

FIG . 16 Patent Application Publication Sep . 20 , 2018 Sheet 18 of 20 US 2018 / 0268942 A1

Precentral Gyrus Postcentral Gyrus ide *14 o Normal $ AD .

S . StoBugle searaO {3? & ORAOND 3000 30

en - -

20 S Inc . 50 60 70 80 90 100 60 70 80 90 100 Age Temporal Lobe Paracentral Lobule 100: 80 o O. Dup asalute 0. # 0000 l mo o - *0° ONONAa . o PNOvalue * . Of O Go d ...... 50 60 70 80 90 100 60To 70 80 90 100 Age Age Olfactory Bulb 80 T Anterior- mid Cingulum

8 ite ano * 00 a ** 55723 ** PNDvalue PNOvalue ** C * 4+ OO + o 50 60 70 80 90 100 50 60 70 80 90 100 Age Age

FIG . 17 Patent Application Publication Sep . 20 , 2018 Sheet 19 of 20 US 2018 / 0268942 A1

ADNIoooo Prediction WIwith Machine Learning¢wa Percent oöe Sensitivity Specificity Positive Negative Accuracy Predictive Predictive is eenValue geregistrerenValue

FIG . 18 Patent Application Publication Sep . 20 , 2018 Sheet 20 of 20 US 2018 / 0268942 A1

00 - zzzz 0 * Normal TO I 0 0 ö 0 O AD 0 0 0 0 80 on 02000000.0 ook 0000 00 o o 000 oo Do00 I 000 0000 ooooo 90# 8600oo San ## 200°00#* . OOO # ## ## *** $ . $ el : ** * & 9 . # . ## po . ------. SA Age

FIG . 19 US 2018 /0268942 A1 Sep . 20 , 2018

METHODS AND SYSTEMS FOR essary for eliminating toxic residue buildup , a process that IDENTIFYING BRAIN DISORDERS can be altered in the AD brain . Yet , there remains limited understanding of brain structural content and its impact on CROSS -REFERENCE transport of molecules in the brain interstitium . Currently , the clinical use and the diagnostic capacity of brain MRI [0001 ] This application claims the benefit of U . S . Provi remains limited to differential diagnosis , only after symp sional Patent Application Ser. No . 62 / 430 , 351, filed Dec. 6 , tomatic presentation , principally due to the inherently low 2016 , and U . S . Provisional Patent Application Ser. No . spatial resolution — MRI image voxels are in mm dimen 62 /481 ,839 , filed Apr. 5 , 2017 , each of which is entirely sions , whereas structural changes contributing to tissue incorporated herein by reference for all purposes . degeneration originate at the sub -micron scale . FDG , amy BACKGROUND loid , and tau PET scans suffer from similar limitations. [ 0002 ] Neurodegenerative diseases leading to dementia SUMMARY are a tremendous societal burden , currently devastating 9 [ 0005 ] Recognized herein is a need for tools that allow million people domestically and 47 million people world early detection of Alzheimer ' s disease and other neurode wide . Current inability to effectively prevent, diagnose and generative disorders , including tools that may utilize combat neurodegeneration results in staggering direct and approaches that detect microscopic changes in brain tissue indirect costs Alzheimer ' s disease ( AD ) , the most common from low -resolution magnetic resonance imaging (MRI ) cause of dementia , alone afflicts over 5 million Americans scans. Such approaches may leverage a deeper understand and accounts for the 6th leading cause of death in the USA . ing of brain tissue microstructure to more reliably predict AD requires an estimated 18 billion hours of unpaid care and interpret the health of the brain from MRI scans well taking and well over $ 250 billion of medical costs annually . before severe tissue damage irreversibly impedes healthy Prevalence of the disease is projected to escalate to nearly 14 cognitive function . million people domestically and 135 million worldwide by [0006 ] Provided herein is an image analysis platform that 2050 , with no potential cure in immediate sight. There is a can detect and quantify brain tissue abnormality ( such as dire need for technological advancements toward diagnos neurodegeneration ) in every voxel of standard clinical brain tics, prevention , therapeutics and eventual cures that will MRI. The platform may provide detailed information about each have profound beneficial impacts on the population . the brain tissue health at the microscopic level and the [0003 ] Current clinical evaluation typically includes non resulting observed patterns of pathologic involvement that is invasive brain imaging with magnetic resonance imaging currently missing in the neuroimaging /brain diagnostics (MRI ) , positron emission tomography ( PET ) , or other field . As a result , complex brain diseases, such as Alzheim advanced imaging strategies which provide insight into er ' s disease, which are currently diagnosed very late ( i. e ., at tissue volume changes , chemical composition , cortical the late symptomatic stages ) , may be diagnosed or otherwise metabolic rate , alterations associated with tissue cellularity identified prior to the onset of advanced symptoms. The and disease biomarkers , and structural abnormalities attrib platform may allow early - stage testing of novel drug can uted to neurodegenerative disease . To aid in the diagnosis of didates for clinical trials , which have previously failed due AD and differential diagnosis from non - Alzheimer demen to poor patient selection , late intervention , and very high tias , fluorodeoxyglucose (FDG ) PET and amyloid PET trial costs . All of these factors may be significantly improved reveal AD - associated patterns of cerebral cortical metabo using the platform . lism and beta amyloid deposits in the gray matter, respec 10007 ] Provided herein are methods and systems for deter tively. Similarly , tau PET reveals neurofibrillary tangles in mining whether brain tissue is indicative of a disorder, such the brain . However, due to a lack of advancement in analysis as a neurodegenerative disorder. The methods and systems technologies , meaningful use of these imaging techniques may allow the early diagnosis of a brain disorder much for neurodegenerative disease is restricted to late stages earlier than would be possible using prior methods and when considerable tissue damage and cognitive or other systems, such as many years before the development of clinical abnormalities are present . As we deepen our under symptoms associated with the disorder that are detectable standing of the multiplicity of abnormalities associated with using prior methods and systems. The methods and systems AD , there is increasing evidence that the continual targeting may provide high accuracy in diagnosing a brain disorder of these amyloid plaques and neurofibrillary tangles may ( such as greater than 90 % accuracy ) , as measured by a merely be treating late stage symptoms rather than the variety of criteria described herein . underlying causes. The inability to effectively detect early 0008 ]. The methods and systems of the present disclosure stages of AD precludes pre -symptomatic intervention and may utilize data processing techniques to assess a level of conceals the potential beneficial effects of drug candidates . congruence between measured parameters obtained from [0004 ] Implicit to the neurodegenerative process is the magnetic resonance imaging (MRI ) data and simulated death of the signaling nerve cells in the brain , though this parameters obtained from computational modeling of brain can merely be the ultimate consequence in a cascade of tissues . The methods and systems generally operate by degeneration within brain tissue . The structural integrity of determining a level of congruence between the one or more tissue is necessary for neuron support and survival and measured parameters and the one or more simulated param clearance of molecular waste thatmust be removed from the eters for one or more voxels of the MRI data . The simulated brain for maintenance of neural tissue homeostasis and parameters are obtained from a plurality of microstructural efficient function . Alterations in non - cellular components of models . Each microstructural model of the plurality of the brain are complicit in the degenerative process and may microstructural models is obtained by subjecting a micro be a precursor of lost nerve cell function . It has been shown structuralmodel that is not indicative of a disorder to a series that proper regulation of neural tissue homeostasis is nec - ofmicrostructural perturbations. After assessing the level of US 2018 /0268942 A1 Sep . 20 , 2018 congruence between the one or more measured parameters of: a longitudinal relaxation time ( T1 ), a transverse relax and the one or more simulated parameters for a number of ation time ( T2 ) , and a diffusion coefficient. microstructural models of the plurality of microstructural f0014 ] The one or more microstructural models may com models , a diagnostic microstructural model that meets a prise information regarding a parameter selected from the threshold congruence is selected . The diagnostic microstruc group consisting of: intracellular content, extracellular con tural model is used to determine the disorder state of the tent, distribution of extracellular content within interstitial brain tissue associated with the voxel. space , distribution of intracellular content within intracellu 10009] The methods and systems may be applied to a lar space , and tissue geometry . The one or more microstruc plurality of voxels of the MRI data , such that a level of tural models may comprise measured or predicted values of congruence is determined for each voxel of the plurality of a parameter selected from the group consisting of : cell voxels . In this manner, a diagnostic model and a disorder density , cell shape , cell geometry , cell size , cell distribution , state may be determined for each voxel of the plurality of intercellular spacing, extracellular matrix homogeneity , voxels . The methods and systemsmay be applied to deter interstitial tortuosity , water to protein ratio , water to lipid mine a diagnostic model and a disorder state for a plurality ratio , water to carbohydrate ratio , protein to lipid ratio , of voxels located within a particular region of a brain , within protein to carbohydrate ratio , and lipid to carbohydrate ratio . a whole brain , or across a plurality of brains from a plurality The one or more microstructural models may be selected of subjects . from a microstructural model library. The microstructural [0010 ] In an aspect , a method for determining a disorder model library may comprise at least 100 microstructural state of brain tissue in a brain of a subject may comprise : ( a ) models . obtaining magnetic resonance imaging (MRI ) data compris [ 0015 ] The microstructural model library may be con ing at least one MRI image of the brain , the MRI image structed by : ( a ) creating a first microstructural model cor comprising a plurality of voxels , a voxel of the plurality of responding to a brain state that is not associated with a voxels being associated with the brain tissue of the brain of disorder ; and ( b ) iteratively subjecting the first microstruc the subject and comprising one or more measured MRI tural model to a perturbation , each iteration producing an parameters in the MRI data ; ( b ) for the voxel of the plurality additional perturbed microstructural model . ( b ) may com of voxels, using one or more computer processors to process prise subjecting the first microstructural model to at least the one or more measured MRI parameters with one or more 100 iterations to generate at least 100 perturbed microstruc simulated MRI parameters for the voxel, the one or more tural models . The first microstructural model may be simulated MRI parameters being generated from one or selected based on knowledge of the brain region associated more microstructural models at the voxel; ( c ) for the voxel with the voxel . The perturbation may comprise an operation of the plurality of voxels , selecting a diagnostic model from selected from the group consisting of: depleting cells, alter the one or more microstructural models , the diagnostic ing cellular morphology or distribution , altering intracellular model meeting a threshold congruence between the one or or interstitial physico - chemical composition or distribution , more measured MRI parameters and the one or more simu altering extracellularmatrix composition or distribution , and lated MRI parameters associated with the diagnostic model; altering intercellular spacing. The perturbation may com and (d ) for the voxel of the plurality of voxels , using the prise a stochastic procedure . diagnostic model to determine the disorder state of the brain [0016 ) The threshold congruence may be determined by tissue associated with the voxel . computing an objective function between the one or more [ 0011 ] Each voxel may comprise a plurality of measured measured MRI parameters and the one or more simulated MRI parameters . The one or more measured MRI param MRI parameters . The objective function may comprise an eters may be a plurality of measured MRI parameters . The L1 norm or an L2 norm . one or more simulated MRI parameter may be a plurality of [00171 Determining the disorder state of the brain tissue simulated MRI parameters . associated with the voxel may be achieved at an accuracy of [0012 ] Themethod may further comprise repeating ( b ) - ( d ) at least 90 % . Determining the disorder state across the brain one or more times for additional voxels of the plurality of tissue associated with the specified region of the brain may voxels . The method may further comprise repeating ( b ) - ( d ) be achieved at an accuracy of at least 90 % . Determining the for all other voxels of the plurality of voxels . The method disorder state of the brain tissue associated with the whole may further comprise repeating ( b ) - ( d ) for all voxels asso brain of the subject may be achieved at an accuracy of at ciated with a specified region of the brain . The method may least 90 % . Determining the disorder state of the brain tissue further comprise repeating (b ) -( d ) for all voxels associated associated with the plurality of subjects may be achieved at with an entirety of the brain . The method may further an accuracy of at least 90 % . comprise repeating (a )- ( d ) for a plurality of MRI images , [ 0018 ] The disorder may be a non - neurodegenerative dis each MRI image of the plurality of MRI images associated order. The disorder may be selected from the group consist with a brain selected from a plurality of brains , each brain ing of: a primary neoplasm , a metastatic neoplasm , a seizure of the plurality of brains associated with a subject selected disorder, a seizure disorder with focal cortical dysplasia , a from a plurality of subjects . demyelinating disorder , a non -neurodegenerative encepha [0013 ] The MRI image may be selected from the group lopathy, a cerebrovascular disease , and a psychological consisting of: a longitudinal relaxation time ( T1 ) -weighted disorder . The disorder may be a neurodegenerative disorder . MRI image, a transverse relaxation time ( T2 ) - weighted MRI The disorder may be selected from the group consisting of: image, and a diffusion - weighted MRI image. The measured Alzheimer' s disease , a non - Alzheimer 's dementia disorder, MRI parameter may be selected from the group consisting Parkinson ' s disease , a Parkinsonism disorder , a motor neu of: a longitudinal relaxation time ( T1 ) , a transverse relax ron disease , Huntington ' s disease , a Huntington ' s disease ation time ( T2 ) , and a diffusion coefficient. The simulated like syndrome, transmissible spongiform encephalopathy , MRI parameter may be selected from the group consisting chronic traumatic encephalopathy, and a tauopathy . US 2018 /0268942 A1 Sep . 20 , 2018

[0019 ] The method may enable diagnosis of a neurode one or more microstructural models , the diagnostic model generative disorder more than 5 years prior to the develop meeting a threshold congruence between the one or more ment of symptoms associated with the neurodegenerative measured MRI parameters and the one or more simulated disorder. The method may enable monitoring of the neuro MRI parameters associated with the diagnostic model; and degenerative disorder at a plurality of time points , the ( d ) for the voxel of the plurality of voxels , using the plurality of time points separated by a plurality of time diagnostic model to determine the disorder state of the brain intervals. tissue associated with the voxel . [0020 ] The method may further comprise constructing a [0024 ] Each voxel may comprise a plurality of measured brain map that, for each voxel of the plurality of voxels , MRI parameters . The one or more measured MRI param indicates the disorder state of the brain tissue associated with eters may be a plurality of measured MRI parameters . The the voxel. The method may further comprise displaying the one or more simulated MRI parameter may be a plurality of brain map on a graphical user interface of an electronic simulated MRI parameters . device of a user. The brain map may comprise a qualitative [0025 ] The method may further comprise repeating (b )- ( d ) abnormality map . The brain map may comprise a binary one or more times for additional voxels of the plurality of abnormality map . The brain map may comprise a quantita voxels . The method may further comprise repeating ( b )- ( d ) tive abnormality map . The brain map may comprise a for all other voxels of the plurality of voxels . The method percent abnormality map . may further comprise repeating ( b ) - ( d ) for all voxels asso [ 0021] In an aspect, a method for determining a disorder ciated with a specified region of the brain . Themethod may state of a tissue in a portion of a body of a subject may further comprise repeating ( b ) - ( d ) for all voxels associated comprise : obtaining magnetic resonance imaging (MRI ) with an entirety of the brain . The method may further data comprising at least one MRI image of the tissue, the comprise repeating (a )- ( d ) for a plurality of MRI images, MRI image comprising a plurality of voxels , a voxel of the each MRI image of the plurality of MRI images associated plurality of voxels being associated with the tissue of the with a brain selected from a plurality of brains , each brain subject and comprising one or more measured MRI param of the plurality of brains associated with a subject selected eters in the MRI data ; (b ) for the voxel of the plurality of from a plurality of subjects . voxels , using one or more computer processors to process [0026 ] The MRI image may be selected from the group the one or more measured MRI parameters with one or more consisting of: a longitudinal relaxation time ( T1 ) - weighted simulated MRI parameters for the voxel, the one or more MRI image , a transverse relaxation time ( T2 ) -weighted MRI simulated MRI parameters being generated from one or image, and a diffusion - weighted MRI image . The measured more microstructural models at the voxel; ( c ) for the voxel MRI parameter may be selected from the group consisting of the plurality of voxels , selecting a diagnostic model from of: a longitudinal relaxation time ( T1 ), a transverse relax the one or more microstructural models , the diagnostic ation time ( T2 ) , and a diffusion coefficient. The simulated model meeting a threshold congruence between the one or MRI parameter may be selected from the group consisting more measured MRI parameters and the one or more simu - of: a longitudinal relaxation time ( T1 ), a transverse relax lated MRI parameters associated with the diagnostic model; ation time ( T2 ) , and a diffusion coefficient. and ( d ) for the voxel of the plurality of voxels, using the [0027 ] The one or more microstructural models may com diagnostic model to determine the disorder state of the tissue prise information regarding a parameter selected from the associated with the voxel . group consisting of: intracellular content, extracellular con [0022 ] The tissue may be selected from the group con tent, distribution of extracellular content within interstitial sisting of: spinal cord tissue, heart tissue, vascular tissue , space , distribution of intracellular content within intracellu lung tissue , liver tissue , kidney tissue , esophageal tissue , lar space , and tissue geometry . The one or more microstruc stomach tissue , intestinal tissue, pancreatic tissue , thyroid tural models may comprise measured or predicted values of tissue , adrenal tissue , spleen tissue , lymphatic tissue , appen a parameter selected from the group consisting of: cell dix tissue , breast tissue, bladder tissue , vaginal tissue , ovar density , cell shape , cell geometry , cell size , cell distribution , ian tissue , uterine tissue , penile tissue , testicular tissue , intercellular spacing , extracellular matrix homogeneity, prostatic tissue , skeletal muscle tissue, skin , and non -brain interstitial tortuosity , water to protein ratio , water to lipid tissue of the head and neck . ratio , water to carbohydrate ratio , protein to lipid ratio , [ 0023 ] In an aspect, a non - transitory computer - readable protein to carbohydrate ratio , and lipid to carbohydrate ratio . medium may comprise machine - executable code that , upon The one or more microstructural models may be selected execution by one or more computer processors, implements from a microstructural model library . The microstructural a method for detecting a disorder state of brain tissue in a model library may comprise at least 100 microstructural brain of a subject, the method comprising : ( a ) obtaining models . magnetic resonance imaging (MRI ) data comprising at least [ 0028 ] The microstructural model library may be con one MRI image of the brain , the MRI image comprising a structed by : ( a ) creating a first microstructural model cor plurality of voxels , a voxel of the plurality of voxels being responding to a brain state that is not associated with a associated with the brain tissue of the brain of the subject disorder; and ( b ) iteratively subjecting the first microstruc and comprising one or more measured MRI parameters in tural model to a perturbation , each iteration producing an the MRI data ; ( b ) for the voxel of the plurality of voxels , additional perturbed microstructural model. ( b ) may com using one or more computer processors to process the one or prise subjecting the first microstructural model to at least more measured MRI parameters with one or more simulated 100 iterations to generate at least 100 perturbed microstruc MRI parameters for the voxel, the one or more simulated tural models . The first microstructural model may be MRI parameters being generated from one or more micro selected based on knowledge of the brain region associated structural models at the voxel; ( c ) for the voxel of the with the voxel. The perturbation may comprise an operation plurality of voxels , selecting a diagnostic model from the selected from the group consisting of: depleting cells, alter US 2018 /0268942 A1 Sep . 20 , 2018 ing cellular morphology or distribution , altering intracellular eters being generated from one or more microstructural or interstitial physico - chemical composition or distribution , models at the voxel ; ( c ) for the voxel of the plurality of altering extracellular matrix composition or distribution , and voxels , selecting a diagnostic model from the one or more altering intercellular spacing . The perturbation may com microstructural models , the diagnostic model meeting a prise a stochastic procedure . threshold congruence between the one or more measured [ 0029 ] The threshold congruence may be determined by MRIparameters and the one or more simulated MRI param computing an objective function between the one or more eters associated with the diagnostic model; and ( d ) for the measured MRI parameters and the one or more simulated voxel of the plurality of voxels , using the diagnostic model MRI parameters . The objective function may comprise an to determine the disorder state of the tissue associated with L1 norm or an L2 norm . the voxel. [0030 ] Determining the disorder state of the brain tissue [0035 ] The tissue may be selected from the group con associated with the voxel may be achieved at an accuracy of sisting of: spinal cord tissue , heart tissue , vascular tissue , at least 90 % . Determining the disorder state across the brain lung tissue , liver tissue , kidney tissue , esophageal tissue , tissue associated with the specified region of the brain may stomach tissue , intestinal tissue , pancreatic tissue , thyroid be achieved at an accuracy of at least 90 % . Determining the tissue , adrenal tissue , spleen tissue, lymphatic tissue , appen disorder state of the brain tissue associated with the whole dix tissue, breast tissue , bladder tissue, vaginal tissue , ovar brain of the subject may be achieved at an accuracy of at ian tissue , uterine tissue , penile tissue , testicular tissue , least 90 % . Determining the disorder state of the brain tissue prostatic tissue , skeletal muscle tissue , skin , and non -brain associated the plurality of subjects may be achieved at an tissue of the head and neck . accuracy of at least 90 % . [ 0036 ] In an aspect , a system for determining a disorder [ 0031 ] The disorder may be a non - neurodegenerative dis state of brain tissue in a brain of a subject may comprise: ( a ) order. The disorder may be selected from the group consist a database comprising magnetic resonance imaging (MRI ) ing of: a primary neoplasm , a metastatic neoplasm , a seizure data comprising at least one MRI image of the brain , the disorder, a seizure disorder with focal cortical dysplasia , a MRI image comprising a plurality of voxels , a voxel of the demyelinating disorder, a non -neurodegenerative encepha plurality of voxels being associated with the brain tissue of lopathy , a cerebrovascular disease , and a psychological the brain of the subject and comprising a measured MRI disorder. The disorder may be a neurodegenerative disorder. parameter in the MRI data ; and (b ) one or more computer The disorder may be selected from the group consisting of: processors operatively coupled to the database , wherein the Alzheimer' s disease , a non - Alzheimer' s dementia disorder , one or more computer processors are individually or col Parkinson ' s disease , a Parkinsonism disorder, a motor neu lectively programmed to : ( i ) for the voxel of the plurality of ron disease, Huntington ' s disease , a Huntington ' s disease voxels , use one or more computer processors to process the like syndrome, transmissible spongiform encephalopathy , one or more measured MRI parameters with one or more chronic traumatic encephalopathy, and a tauopathy . simulated MRI parameters for the voxel, the one or more [ 0032 ] The method may enable diagnosis of a neurode simulated MRI parameters being generated from one or generative disorder more than 5 years prior to the develop more microstructural models at the voxel; ( ii ) for the voxel ment of symptoms associated with the neurodegenerative of the plurality of voxels , select a diagnostic model from the disorder . Themethod may enable monitoring of the neuro one or more microstructural models , the diagnostic model degenerative disorder at a plurality of time points , the meeting a threshold congruence between the one or more plurality of time points separated by a plurality of time measured MRI parameters and the one or more simulated intervals . MRI parameters associated with the diagnostic model; and [0033 ] The method may further comprise constructing a ( iii ) for the voxel of the plurality of voxels, use the diag brain map that, for each voxel of the plurality of voxels , nostic model to determine the disorder state of the brain indicates the disorder state of the brain tissue associated with tissue associated with the voxel. the voxel. The method may further comprise displaying the [0037 ] Each voxel may comprise a plurality of measured brain map on a graphical user interface of an electronic MRI parameters . The one or more measured MRI param device of a user . The brain map may comprise a qualitative eters may be a plurality of measured MRI parameters . The abnormality map . The brain map may comprise a binary one ormore simulated MRI parameter may be a plurality of abnormality map . The brain map may comprise a quantita simulated MRI parameters . tive abnormality map . The brain map may comprise a [0038 ] The one or more computer processors may be percent abnormality map . further individually or collectively programmed to repeat [0034 ] In an aspect , a non - transitory computer - readable ( i ) - ( iii ) one or more times for additional voxels of the medium may comprise machine - executable code that, upon plurality of voxels . The one or more computer processors execution by one or more computer processors, implements may be further individually or collectively programmed to a method for detecting a disorder state of brain tissue in a repeat (i )- ( iii ) for all other voxels of the plurality of voxels . brain of a subject , the method comprising : obtaining mag The one or more computer processors may be further netic resonance imaging (MRI ) data comprising at least one individually or collectively programmed to repeat ( 1 ) - ( iii ) for MRI image of the tissue , the MRI image comprising a all voxels associated with a specified region of the brain . The plurality of voxels , a voxel of the plurality of voxels being one or more computer processors may be further individu associated with the tissue of the subject and comprising one ally or collectively programmed to repeat ( i ) - ( iii) for all or more measured MRI parameters in the MRI data ; ( b ) for voxels associated with an entirety of the brain . The one or the voxel of the plurality of voxels, using one or more more computer processors may be further individually or computer processors to process the one or more measured collectively programmed to repeat (i ) - ( iii ) for a plurality of MRI parameters with one or more simulated MRI param MRI images , each MRI image of the plurality of MRI eters for the voxel, the one or more simulated MRI param - images associated with a brain selected from a plurality of US 2018 /0268942 A1 Sep . 20, 2018 brains , each brain of the plurality of brains associated with disorder, a seizure disorder with focal cortical dysplasia , a a subject selected from a plurality of subjects . demyelinating disorder, a non -neurodegenerative encepha [0039 ] The MRI image may be selected from the group lopathy , a cerebrovascular disease , and a psychological consisting of: a longitudinal relaxation time ( T1 ) -weighted disorder . The disorder may be a neurodegenerative disorder. MRI image, a transverse relaxation time ( T2 ) -weighted MRI The disorder may be selected from the group consisting of: image , and a diffusion - weighted MRI image. The measured Alzheimer ' s disease , a non - Alzheimer ' s dementia disorder, MRI parameter may be selected from the group consisting Parkinson ' s disease , a Parkinsonism disorder , a motor neu of : a longitudinal relaxation time ( T1 ) , a transverse relax ron disease, Huntington 's disease, a Huntington ' s disease ation time ( T2 ), and a diffusion coefficient. The simulated like syndrome, transmissible spongiform encephalopathy , MRI parameter may be selected from the group consisting chronic traumatic encephalopathy , and a tauopathy . of: a longitudinal relaxation time ( T1 ) , a transverse relax 10045 ] The system may enable diagnosis of a neurodegen ation time ( T2) , and a diffusion coefficient . erative disorder more than 5 years prior to the development [ 0040 ] The one or more microstructuralmodels may com of symptoms associated with the neurodegenerative disor prise information regarding a parameter selected from the der. The system may enable monitoring of the neurodegen group consisting of: intracellular content, extracellular con erative disorder at a plurality of time points , the plurality of tent, distribution of extracellular content within interstitial time points separated by a plurality of time intervals . space, distribution of intracellular content within intracellu [0046 ] The one or more computer processors may be lar space , and tissue geometry . The one or more microstruc further individually or collectively programmed to construct tural models may comprise measured or predicted values of a brain map that, for each voxel of the plurality of voxels , a parameter selected from the group consisting of: cell indicates the disorder state of the brain tissue associated with density , cell shape , cell geometry , cell size , cell distribution , the voxel. The one or more computer processors may be intercellular spacing, extracellular matrix homogeneity , further individually or collectively programmed to display interstitial tortuosity , water to protein ratio , water to lipid the brain map on a graphical user interface of an electronic ratio , water to carbohydrate ratio , protein to lipid ratio , device of a user . The brain map may comprise a qualitative protein to carbohydrate ratio , and lipid to carbohydrate ratio . abnormality map . The brain map may comprise a binary The one or more microstructural models may be selected abnormality map . The brain map may comprise a quantita from a microstructural model library . The microstructural tive abnormality map . The brain map may comprise a model library may comprise at least 100 microstructural percent abnormality map . models . [0047 ] In an aspect, a system for determining a disorder [0041 ] The microstructural model library may be con state of a tissue in a portion of a body of a subject may structed by : (a ) creating a first microstructural model cor comprise : ( a ) a database comprising magnetic resonance responding to a brain state that is not associated with a imaging (MRI ) data comprising at least one MRI image of disorder ; and ( b ) iteratively subjecting the first microstruc the brain , the MRI image comprising a plurality of voxels , tural model to a perturbation , each iteration producing an a voxel of the plurality of voxels being associated with the additional perturbed microstructural model. ( b ) may com brain tissue of the brain of the subject and comprising a prise subjecting the first microstructural model to at least measured MRI parameter in the MRI data ; and (b ) one or 100 iterations to generate at least 100 perturbed microstruc more computer processors operatively coupled to the data tural models . The first microstructural model may be base , wherein the one or more computer processors are selected based on knowledge of the brain region associated individually or collectively programmed to : ( i ) for the voxel with the voxel. The perturbation may comprise an operation of the plurality of voxels , use one or more computer pro selected from the group consisting of: depleting cells , alter cessors to process the one or more measured MRI param ing cellular morphology or distribution , altering intracellular eters with one or more simulated MRI parameters for the or interstitial physico - chemical composition or distribution , voxel, the one or more simulated MRI parameters being altering extracellular matrix composition or distribution , and generated from one or more microstructural models at the altering intercellular spacing. The perturbation may com voxel; ( ii ) for the voxel of the plurality of voxels , select a prise a stochastic procedure . diagnostic model from the one or more microstructural [ 0042 ] The threshold congruence may be determined by models , the diagnostic model meeting a threshold congru computing an objective function between the one or more ence between the one or more measured MRI parameters measured MRI parameters and the one or more simulated and the one or more simulated MRI parameters associated MRI parameters . The objective function may comprise an with the diagnostic model; and ( iii) for the voxel of the Li norm or an L2 norm . plurality of voxels , use the diagnostic model to determine [0043 ] Determining the disorder state of the brain tissue the disorder state of the tissue associated with the voxel. associated with the voxel may be achieved at an accuracy of [0048 ] The tissue may be selected from the group con at least 90 % . Determining the disorder state across the brain sisting of: spinal cord tissue , heart tissue, vascular tissue , tissue associated with the specified region of the brain may lung tissue, liver tissue , kidney tissue , esophageal tissue , be achieved at an accuracy of at least 90 % . Determining the stomach tissue , intestinal tissue , pancreatic tissue, thyroid disorder state of the brain tissue associated with the whole tissue , adrenal tissue, spleen tissue , lymphatic tissue , appen brain of the subject may be achieved at an accuracy of at dix tissue, breast tissue , bladder tissue, vaginal tissue , ovar least 90 % . Determining the disorder state of the brain tissue ian tissue, uterine tissue , penile tissue , testicular tissue , associated the plurality of subjects may be achieved at an prostatic tissue , skeletal muscle tissue , skin , and non -brain accuracy of at least 90 % . tissue of the head and neck . [ 0044 ] The disorder may be a non -neurodegenerative dis [0049 ] Additional aspects and advantages of the present order . The disorder may be selected from the group consist disclosure will become readily apparent to those skilled in ing of: a primary neoplasm , a metastatic neoplasm , a seizure this art from the following detailed description , wherein US 2018 /0268942 A1 Sep . 20 , 2018 only illustrative embodiments of the present disclosure are [0064 ] FIG . 11 shows an early detection analysis of MRI shown and described . As will be realized , the present images collected longitudinally from the population distri disclosure is capable of other and different embodiments , bution . and its several details are capable of modifications in various [0065 ] FIG . 12 shows a determination of abnormality in a obvious respects , all without departing from the disclosure . mixed cohort longitudinal study . Accordingly , the drawings and description are to be regarded [0066 ] FIG . 13 shows the registration or alignment of as illustrative in nature , and not as restrictive . subject images to an annotated parcellation atlas . INCORPORATION BY REFERENCE [0067 ] FIG . 14 shows optimized single region prediction accuracy of diagnosis within the ADNI dataset for a variety [0050 ] All publications , patents , and patent applications of brain regions using the systems and methods described mentioned in this specification are herein incorporated by herein . reference to the same extent as if each individual publica [0068 ] FIG . 15 shows PND measurement distributions tion , patent, or patent application was specifically and indi across subjects of a variety of ages for the whole brain , vidually indicated to be incorporated by reference . To the cerebellum , thalamus , posterior cingulate , precuneus, and extent publications and patents or patent applications incor hippocampus . porated by reference contradict the disclosure contained in [0069 ] FIG . 16 shows PND measurement distributions the specification , the specification is intended to supersede across subjects of a variety of ages for the , and / or take precedence over any such contradictory mate basal ganglia , parietal lobe, occipital lobe, prefrontal cortex , rial . and premotor cortex . [0070 ] FIG . 17 shows PND measurement distributions BRIEF DESCRIPTION OF THE DRAWINGS across subjects of a variety of ages for the precentral gyrus, [ 0051] The novel features of the invention are set forth postcentral gyrus, temporal lobe , paracentral lobule , olfac with particularity in the appended claims. A better under tory bulb , and anterior -mid cingulum . standing of the features and advantages of the present [0071 ] FIG . 18 shows attainable AD diagnostic metrics invention will be obtained by reference to the following using machine learning . detailed description that sets forth illustrative embodiments , 100721. FIG . 19 shows a distribution of whole brain scores in which the principles of the invention are utilized , and the (WBS ) for ADNI subject scans. accompanying drawings ( also “ Figure ” and “ FIG .” herein ) , of which : DETAILED DESCRIPTION 10052 ] FIG . 1 shows a method for determining a neuro [0073 ] While various embodiments of the invention are logical disorder state of brain tissue in a brain of a subject. shown and described herein , it will be obvious to those [0053 ] FIG . 2 shows a method for constructing a micro skilled in the art that such embodiments are provided by way structuralmodel library . of example only . Numerous variations, changes, and substi [0054 ] FIG . 3 shows a method for determining a neuro tutions may occur to those skilled in the art without depart ing from the invention . It should be understood that various logical disorder state of brain tissue in a brain of a subject . alternatives to the embodiments of the invention described 10055 ] FIG . 4 shows a computer system that is pro herein may be employed . grammed or otherwise configured to operate a system or 10074 ]. Where values are described as ranges , it will be method for determining a disorder state of brain tissue in a understood that such disclosure includes the disclosure of all subject. possible sub - ranges within such ranges , as well as specific [0056 ] FIG . 5A shows a portion of a cell array for a normal numerical values that fall within such ranges irrespective of tissue microstructural model. whether a specific numerical value or specific sub -range is [0057 ] FIG . 5B shows a sample Monte Carlo simulation expressly stated . with delivery of freely moving molecules. [0075 ] As used herein , the term “ subject” generally refers [0058 ] FIG . 5C shows representative models of healthy to an animal, such as a mammalian species ( e . g . , human ) or and degenerating brain tissue . avian ( e . g . , bird ) species , or other organism , such as a plant. [0059 ] FIG . 6 shows the processing of human magnetic The subject can be a vertebrate , a mammal, a mouse , a resonance imaging (MRI ) images to produce neurodegen primate , a simian , or a human . Animals may include , but are eration maps. not limited to , farm animals , sport animals , and pets . A subject can be a healthy or asymptomatic individual, an [0060 ] FIG . 7 shows examples of MRI, percent neurode individual that has or is suspected of having a disease ( e . g . , generation (PND ) , and quantitative neurodegeneration a neurological disorder) or a pre -disposition to the disease , ( QND ) brain maps from a diseased individual . or an individual that is in need of therapy or suspected of [0061 ] FIG . 8 shows exemplary brain maps for a young , needing therapy . A subject can be a patient. asymptomatic brain , a normal aged brain with no detected [00761 As used herein , the term “ brain region ” (also neurodegenerative symptoms, and an aged brain with clini referred to as “ region of a brain ” or “ region of the brain ” ) cal symptoms of severe neurodegeneration . generally refers to any sub - structure of a brain . The brain [ 0062 ] FIG . 9 shows mean plots of degenerative analysis region may be a sub - region or the entirety of a prosencepha output parameters for Alzheimer ' s Disease Neuroimaging lon ( ) , or a sub -region or the entirety of a mesen Initiative (ADNI ) images . cephalon ( ), or a sub - region or the entirety of a [0063 ] FIG . 10 shows a population distribution for a rhombencephalon ( hindbrain ) . The brain region may be a longitudinal study to evaluate early detection of Alzheimer ' s . The brain region may be a medullary disease ( AD ). pyramid , olivary body , inferior olivary nucleus , rostral ven US 2018 /0268942 A1 Sep . 20 , 2018 trolateral medulla , caudal ventrolateral medulla , solitary posterior nucleus, ventral posterior lateral nucleus , ventral nucleus, respiratory center, dorsal respiratory group, ventral posterior medial nucleus, metathalamus, medial geniculate respiratory group , pre - Botzinger complex , Bötzinger com body, lateral geniculate body, or thalamic reticular nucleus . plex , retrotrapezoid nucleus, nucleus retrofacialis , nucleus The brain region may be a . The brain region retroambiguus, nucleus paraambiguus , paramedian reticular may be an anterior hypothalamus, medial area of the anterior nucleus, gigantocellular reticular nucleus , parafacial zone, hypothalamus, anterior medial , medial preop cuneate nucleus , gracile nucleus , perihypoglossal nucleus , tic nucleus, , paraventricular intercalated nucleus , prepositus nucleus, sublingual nucleus, nucleus, , anterior hypothalamic nucleus , area postrema, medullary cranial nerve nucleus , inferior lateral area of the anterior hypothalamus , anterior lateral salivatory nucleus, nucleus ambiguus, dorsal nucleus of the preoptic area , anterior part of the lateral nucleus, supraoptic vagus nerve, or hypoglossal nucleus. The brain region may nucleus , median preoptic nucleus , periventricular preoptic be a . The brain region may be a pontine nucleus, nucleus , tuberal hypothalamus , medial area of the tuberal pontine cranial nerve nucleus , pontine nucleus of the hypothalamus, dorsomedial hypothalamic nucleus, ventro trigeminal nerve sensory nucleus, motor nucleus for the medial nucleus, , lateral area of the tuberal trigeminal nerve , abducens nucleus, vestibulocochlear hypothalamus , tuberal part of the lateral nucleus , lateral nucleus, superior salivatory nucleus, pontine , tuberal nucleus, posterior hypothalamus, medial area of the pontine micturition center (Barrington ' s nucleus ), locus posterior hypothalamus, mammillary nucleus, posterior coeruleus, pedunculopontine nucleus , laterodorsal tegmen nucleus, lateral area of the posterior hypothalamus, posterior tal nucleus, tegmental pontine reticular nucleus, parabra part of the lateral nucleus, optic chiasm , subfornical organ , chial area , medial parabrachial nucleus, lateral parabrachial periventricular nucleus , , , nucleus, subparabrachial nucleus (Kölliker - Fuse nucleus ) , tuberal nucleus, or tuberomammillary nucleus . The brain pontine respiratory group , superior olivary complex , para region may be a . The brain region may be a median pontine reticular formation , parvocellular reticular or . The brain region may nucleus , caudal pontine reticular nucleus , medial nucleus of be a . The brain region may be a neurohypo the trapezoid body, cerebellar peduncle, superior cerebellar physis , pars intermedia ( intermediate lobe ) , or adenohypo peduncle , middle cerebella peduncle , or inferior cerebellar physis . The brain region may be a cerebrum (telencephalon ) . peduncle . The brain region may be a cerebellum . The brain The brain region may be a , centrum semiovale , region may be a cerebellar vermis , cerebellar hemisphere , corona radiate , internal capsule , external capsule , extreme anterior lobe, posterior lobe, flocculonodular lobe , inter capsule , subcortical cerebrum , hippocampus (medial tem posed nucleus , globose nucleus, emboliform nucleus, or poral lobe ) , , cornu ammonis , cornu ammonis dentate nucleus . The brain region may be a midbrain (mes area 1 , cornu ammonis area 2 , cornu ammonis area 3 , cornu encephalon ) . The brain region may be a tectum . The brain ammonis area 4 , ( limbic lobe ) , central nucleus of region may be a corpora quadrigemina, inferior colliculi, or the amygdala , medial nucleus of the amygdala , cortical superior colliculi . The brain region may be a pretectum . The nucleus of the amygdala , basomedial nucleus of the brain region may be a tegmentum . The brain region may be amygdala , lateral nucleus of the amygdala , basolateral a periaqueductal gray , rostral interstitial nucleus of medial nucleus of the amygdala , , bed nucleus of the longitudinal fasciculus , midbrain reticular formation , dorsal stria terminalis , claustrum , basal ganglia , , , red nucleus , , parabra striatum (neostriatum ), , , ventral chial pigmented nucleus , paranigral nucleus, rostromedial striatum , , , globus tegmental nucleus , caudal linear nucleus , rostral linear pallidus , subthalamic nucleus, basal forebrain , anterior per nucleus of the raphe , interfascicular nucleus, substantia forated substance, substantia innominate , , nigra , pars compact, pars reticulate , or interpeduncular diagonal band of Broca , septal nucleus , . The brain region may be a cerebral peduncle . The nucleus , lamina terminalis , or the vascular organ of the brain region may be a crus cerebri . The brain region may be lamina terminalis . The brain region may be a rhinencephalon a mesencephalic cranial nerve nucleus . The brain region (paleopallium ) . The brain region may be an olfactory bulb , may be an oculomotor nucleus , Edinger - Westphal nucleus , olfactory tract, anterior olfactory nucleus, piriform cortex , or trochlear nucleus . The brain region may be a mesenchep anterior commissure , uncus , or periamygdaloid cortex . The halic duct (aqueduct of Sylvius ). The brain region may be a brain region may be a (neopallium ) . The forebrain (prosencephalon ) . The brain region may be a brain region may be a frontal lobe, frontal lobe cortex , . The brain region may be an . The primary motor cortex ( precentral gyrus) , supplementary brain region may be a pineal body, habenular nucleus, stria motor cortex , premotor cortex , prefrontal cortex , orbitofron medullaris , or taenia thalami. The brain region may be a tal cortex , dorsolateral prefrontal cortex , frontal lobe gyrus , third ventricle . The brain region may be a fourth ventricle . superior frontal gyrus , middle frontal gyrus , inferior frontal The brain region may be a lateral ventricle . The brain region gyrus , paracentral lobule , Brodmann area 4 , Brodmann area may be a . The brain region may be a 6 , Brodmann area 8 , Brodmann area 9 , Brodmann area 10 , thalamus . The brain region may be a anteroventral nucleus , Brodmann area 11 , Brodmann area 12 , Brodmann area 24 , anterodorsal nucleus, anteromedial nucleus , medial nuclear Brodmann area 25 , Brodmann area 32 , Brodmann area 33 , group , , midline nuclear group , parate Brodmann area 44 , Brodmann area 45 , Brodmann area 46 , nial nucleus, reuniens nucleus, rhomboidal nucleus, intrala Brodmann area 47, parietal lobe, parietal lobe cortex , pri minar nuclear group , , parafascicular mary somoatosensory cortex , secondary somatosensory cor nucleus, paracentral nucleus , central lateral nucleus , central tex , posterior parietal cortex , parietal lobe gyrus , postcentral medial nucleus , lateral nuclear group , lateral dorsal nucleus , gyrus , precuneus , posterior cingulate cortex , Brodmann area lateral posterior nucleus , pulvinar, ventral nuclear group , 1 , Brodmann area 2 , Brodmann area 3 , Brodmann area 5 , , ventral lateral nucleus , ventral Brodmann area 7 , Brodmann area 23 , Brodmann area 26 , US 2018 /0268942 A1 Sep . 20 , 2018

Brodmann area 29 , Brodmann area 31 , Brodmann area 39 , tern , cistern of lamina terminalis , chiasmatic cistern , inter Brodmann area 40 , occipital lobe , occipital lobe cortex , peduncular cistern , pontine cistern , cisterna magna , or spinal primary visual cortex , secondary visual cortex , third visual subarachnoid space . The brain region may comprise any 1 , cortex , fourth visual cortex , dorsomedial area , middle tem 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , 20 , 30 , 40 , 50 , 60 , 70 , 80 , 90 , 100 , poral visual cortex , occipital lobe gyrus, lateral occipital 200 , 300 , 400 , 500 , 600 , 700 , 800 , 900 , 1 ,000 , or more than 1 , 000 brain regions described herein . The brain region may gyrus, cuneus, Brodmann area 17 , Brodmann area 18 , Brod comprise a number of brain regions that is within a range mann area 19 , temporal lobe , temporal lobe cortex , primary defined by any two of the preceding values . auditory cortex , secondary auditory cortex, inferior temporal [0077 ] As used herein , the term “ tissue " generally refers to cortex , posterior inferior temporal cortex , temporal lobe biological tissue. The biological tissue may be from a gyrus , superior temporal gyrus , middle temporal gyrus , subject. inferior temporal gyrus , entorhinal cortex , perirhinal cortex , 10078 ] As used herein , the term “ voxel” generally refers to parahippocampal gyrus , fusiform gyrus , Brodmann area 20 , a unit volume in three - dimensional space . The voxel may Brodmann area 21 , Brodmann area 22 , Brodmann area 27 , correspond to a given three - dimensional volume, such as a Brodmann area 34 , Brodmann area 35 , Brodmann area 36 , volume of biological tissue . In the context of biological Brodmann area 37 , Brodmann area 38 , Brodmann area 41 , tissue, a voxel may represent a unit volume of the tissue or Brodmann area 42 , medial superior temporal area , insular a portion of the tissue. For example , in the brain a voxel may cortex , cingulate cortex , anterior cingulate cortex , retrosple represent a unit volume of the brain . Such unit volume may nial cortex , indusium griseum , Brodmann area 23 , Brod be generated by a user; for instance , a usermay generate the mann area 24 , Brodmann area 26 , Brodmann area 29 , unit volume by selecting one or more parameters in a MRI Brodmann area 30 , Brodmann area 31, or Brodmann area pulse sequence . In some cases, such unit volume may relate 32 . The brain region may be a neural pathway . The brain to a biological unit of the tissue . For example , a voxel in the region may be a superior longitudinal fasciculus, arcuate context of the brain may correspond to a neuron , a grouping fasciculus , perforant pathway, thalamocortical radiation , of neurons, one or more brain regions, or one or more corpus callosum , anterior commissure , interthalamic adhe portions of one or more brain regions . A voxel may take a sion , posterior commissure , , , rectilinear form , such as that a cube or rectangular prism . A mammillotegmental fasciculus , cerebral peduncle , medial voxel may be defined by any combination of a first linear forebrain bundle , medial longitudinal fasciculus, myoclonic dimension , a second linear dimension , and a third linear triangle , major dopaminergic pathway , mesocortical path dimension . The first linear dimension may be atmost 10 um , way , , , tuberorin at most 20 um , at most 50 um , at most 100 um , at most 200 fundibular pathway , pathway, raphe nucleus, nor um , at most 500 um , at most 1 mm , at most 2 mm , at most epinephrine pathway , , epinephrine pathway , 5 mm , or at most 10 mm . The first linear dimension may glutamate pathway, or pathway . The brain have a value that is within a range defined by any two of the region may be a descending fiber. The brain region may be preceding values . The second linear dimension may be at an extrapyramidal system , pyramidal tract, corticospinal most 10 um , at most 20 um , at most 50 um , atmost 100 um , tract , lateral corticospinal tract , anterior corticospinal tract, at most 200 um , at most 500 um , at most 1 mm , at most 2 corticopontine fiber, frontopontine fiber, temporopontine mm , at most 5 mm , or at most 10 mm . The second linear fiber, corticobulbar tract, corticomesencephalic tract, tec dimension may have a value that is within a range defined tospinal tract, interstitiospinal tract, rubrospinal tract, by any two of the preceding values . The third linear dimen rubroolivary tract, olivocerebellar tract, olivospinal tract, vestibulospinal tract , lateral vestibulospinal tract, medial sion may be at most 10 um , at most 20 um , at most 50 um , vestibulospinal tract, reticulospinal tract, lateral raphespinal at most 100 um , at most 200 um , at most 500 um , at most tract , alpha system , or gamma system . The brain region may 1 mm , at most 2 mm , at most 5 mm , or at most 10 mm . The be a somatosensory system . The brain region may be a third linear dimension may have a value that is within a posterior column , pathway , gracile fas range defined by any two of the preceding values. ciculus, cuneate fasciculus , medial lemniscus , , lateral spinothalamic tract, anterior spinothalamic Methods for Determining a State of Tissue tract , spinomesencephalic tract , spinocerebellar tract, spi [0079 ] In an aspect , the present disclosure provides meth noolivary tract , or spinoreticular tract . The brain region may ods for determining a state of tissue, such as a disorder state be a visual system . The brain region may be an optic tract or of brain tissue . A method for determining a disorder state of . The brain region may be an auditory system . a tissue in a portion of a body of a subject may comprise The brain region may be a trapezoid body . The brain region obtaining magnetic resonance imaging (MRI ) data compris may be a . The brain region may be a brain ing at least one MRI image of the tissue . The MRI image stem . The brain region may be a cranial nerve , terminal may comprise a plurality of voxels . A voxel of the plurality nerve , olfactory nerve , optic nerve , oculomotor nerve , tro of voxels may be associated with the tissue of the subject chlear nerve , trigeminal nerve , abducens nerve , facial nerve , and comprise one or more measured MRI parameters in the vestibulocochlear nerve , glossopharyngeal nerve, vagus MRI data . nerve, accessory nerve , or hypoglossal nerve . The brain 10080 ) Next , for the voxel of the plurality of voxels , one region may be a neurovascular system . The brain region may or more computer processors may be used to process the one be a middle cerebral artery , posterior cerebral artery , anterior or more measured MRI parameters with one or more simu cerebral artery -vertebral artery , basilar artery , circle of Wil lated MRI parameters for the voxel. The one or more lis , glymphatic system , venous system , or circumventricular simulated MRI parameters may be generated from one or organ . The brain region may be a meningeal covering , dura more microstructural models at the voxel. mater, arachnoid mater, pia mater, epidural space, subdural [ 0081 ] Next, for the voxel of the plurality of voxels , a space , subarachnoid space , arachnoid septum , superior cis diagnostic model may be selected from the one or more US 2018 /0268942 A1 Sep . 20 , 2018 microstructural models. The diagnostic model may be [ 0086 ] One or more of the MRI images may comprise a selected using a threshold congruence between the one or weighted MRI image . One or more MRI images may more measured MRI parameters and the one or more simu comprise a longitudinal relaxation time ( T1 ) -weighted MRI lated MRI parameters associated with the diagnostic model. image . The one or more T1 -weighted MRI images may be For the voxel of the plurality of voxels , the diagnostic model obtained by a T1 -weighted MRI pulse sequence , such as a may be used to determine the disorder state of the tissue T1 -weighted spin echo pulse sequence , a T1 -weighted gra associated with the voxel . dient echo pulse sequence , a paramagnetic contrast agent 10082 ] Methods of the present disclosure may be used to ( such as gadolinium ) enhanced T1 -weighted pulse sequence , determine a disorder state of brain tissue of a subject. A a T1 -weighted Fluid - Attenuated Inversion Recovery ( T1 method for determining a disorder state of brain tissue of a FLAIR ) pulse sequence , a fat- suppressed T1 -weighted pulse subject may comprise obtaining MRI data comprising at sequence , or any other T1 -weighted MRI pulse sequence . least one MRI image of the brain , the MRI image compris One or more MRI images may comprise a transverse relax ing a plurality of voxels , a voxel of the plurality of voxels ation time ( T2 ) -weighted MRI image . The one or more being associated with the brain tissue of the brain of the T2 -weighted MRI images may be obtained by a subject and comprising one or more measured MRI param T2 - weighted MRI pulse sequence , such as a T2 - weighted eters in the MRI data . Next, for the voxel of the plurality of spin echo pulse sequence , a T2 -weighted gradient echo pulse voxels , one or more computer processors may be used to sequence , a T2 -weighted Fluid - Attenuated Inversion Recov process the one or more measured MRI parameters with one ery (T2 -FLAIR ) pulse sequence , a fat -suppressed or more simulated MRI parameters for the voxel. The one or T2 -weighted pulse sequence , a T2 - star pulse sequence, or more simulated MRI parameters may be generated from one any other T2 - weighted MRI pulse sequence . One or more or more microstructural models at the voxel. For the voxel MRI images may comprise a diffusion -weighted MRI of the plurality of voxels , a diagnostic model may be image. The one or more diffusion -weighted MRI images selected from the one or more microstructural models. The may be obtained by any diffusion -weighted MRI pulse diagnostic model may meet a threshold congruence between sequence , such as a diffusion - weighted imaging (DWI ) the one or more measured MRI parameters and the one or pulse sequence , a diffusion tensor imaging (DTI ) pulse more simulated MRI parameters associated with the diag sequence , or a diffusion kurtosis imaging (DKI ) pulse nostic model. Next, for the voxel of the plurality of voxels , sequence . One or more MRI images may comprise a proton the diagnostic model may be used to determine the disorder density (PD ) -weighted MRI image. The one or more proton state of the brain tissue associated with the voxel. density -weighted MRI images may be obtained by any [0083 ] Reference will now be made to the figures, wherein proton density -weighted MRI pulse sequence , such as to a like numerals refer to like parts throughout . It will be fat- suppressed proton density -weighted pulse sequence . One appreciated that the figures and elements therein are not or more MRI images may comprise a post- processed diffu necessarily drawn to scale . sion - weighted image such as an apparent diffusion coeffi cient (ADC ) image , a mean diffusivity (MD ) image , an axial [ 0084 ] FIG . 1 shows a method 100 for determining a diffusivity (AxD ) image, a radial diffusivity (RD ) image, or disorder state of brain tissue in a brain of a subject. a fractional anisotropy ( FA ) image . The one or more post 10085 ). In a first operation 110 , the method may comprise processed diffusion - weighted MRI images may be obtained obtaining magnetic resonance imaging (MRI ) data . The by post -processing of any diffusion -weighted MRI pulse MRI data may comprise MRI data obtained from at least one sequence, such as a diffusion -weighted imaging (DWI ) subject. The MRI data may comprise MRI data obtained pulse sequence , a diffusion tensor imaging (DTI ) pulse from at least 1 , at least 2 , at least 5 , at least 10 , at least 20 , sequence , or a diffusion kurtosis imaging (DKI ) pulse at least 50 , at least 100 , at least 200 , at least 500 , at least sequence . One or more MRI images may comprise a sus 1 ,000 , at least 2 , 000 , at least 5 , 000 , at least 10 , 000 , at least ceptibility -weight image, a spoiled gradient echo (SPGR ) 20 ,000 , at least 50 ,000 , at least 100 , 000 , at least 200 ,000 , at image , a fast spoiled gradient echo (FSPGR ) image, an least 500 , 000 , or at least 1 , 000 ,000 subjects . The MRI data inversion recovery spoiled gradient echo ( IR _ SPGR ) image , may comprise MRI data obtained from a number of subjects a magnetization prepared rapid gradient echo (MP RAGE ) that is within a range defined by any two of the preceding image, or a fluid - attenuated inversion recovery ( FLAIR ) values . TheMRI data may comprise at least one MRI image . image . One or more MRI images may comprise a sodium The MRI data may comprise at least 1 , at least 2 , at least 5 , magnetic resonance (sodium MRI) image , a susceptibility at least 10 , at least 20 , at least 50 , at least 100 , at least 200 , weighted image (SWI ) , a magnetic resonance spectroscopy at least 500 , at least 1 , 000 , at least 2 ,000 , at least 5 ,000 , at (MRS ) image , a magnetic resonance fingerprinting (MRF ) least 10 ,000 , at least 20 , 000 , at least 50 , 000 , at least 100 , 000 , at least 200 ,000 , at least 500 , 000 , or at least 1 , 000 , 000 MRI image , a functional magnetic resonance ( fMRI) image , such images . The MRI data may comprise a number of MRI as a blood - oxygen - level- dependent (BOLD ) image , or an images that is within a range defined by any two of the arterial spin labeling ( ASL ) image. preceding values. The MRI data may comprise a single MRI [ 0087 ] Each MRI image may comprise a plurality of image of each brain of each subject of the plurality of voxels . Each voxel may be associated with brain tissue of subjects . Alternatively , the MRI data may comprise a plu the one or more brains of the one or more subjects . Each rality ofMRI images of each brain of each subject, such as voxel may comprise one or more measured MRI parameters . at least 2 , at least 3 , at least 4 , at least 5 , at least 10 , at least The measured MRI parameters may comprise a measured 20 , at least 50 , at least 100 , at least 200 , at least 500 , or at T1 time, a measured T2 time, a measured proton density , a least 1 , 000 images of each brain of each subject. The number measured diffusion coefficient, a measured diffusivity , a of images for each subject of the plurality of subjects may measured fractional anisotropy of diffusion , or a measured be the same across all subjects . Alternatively , the number of diffusion kurtosis . The measured MRI parameters may com images for each subject may differ across the subjects . prise a plurality of measured MRI parameters . For instance , US 2018 /0268942 A1 Sep . 20 , 2018 10 the measured MRI parameters may comprise a measured Ti the microstructural model parameters for a voxel associated time and a measured T2 time, a measured T1 time and a with a particular region of the brain may be assigned based measured diffusion coefficient, a measured T2 time and a on theoretical predictions of the values of the parameters measured diffusion coefficient, or a measured T1 time, a within the given region . In this manner, the one or more measured T2 time, and a measured diffusion coefficient. The microstructuralmodel parameters may be dependent on the number ofmeasured MRI parameters for each voxelmay be region of the brain with which a voxel is associated , and the the same across all voxels, all images , or all subjects . microstructuralmodel parameters may be different for other Alternatively, the number of measured MRI parameters for voxels associated with different regions of the brain . each voxel may differ across the voxels , images, or subjects . 10093 ] The one or more microstructural models may be [ 0088 ] In a second operation 120, the method may com selected from one or more microstructural model libraries . prise using one or more computer processors to process the Each of the one or more microstructuralmodel libraries may one or more measured MRI parameters for a voxel of the comprise at least 100 , at least 200 , at least 500 , at least plurality of voxels . The one or more measured MRI param 1 , 000 , at least 2 , 000 , at least 5 , 000 , at least 10 , 000 , at least eters may be processed with one or more simulated MRI 20 , 000 , at least 50 , 000 , at least 100 , 000 , at least 200 , 000 , at parameters. least 500 ,000 , or at least 1 , 000 ,000 microstructural models . [0089 ] The simulated MRI parameters may comprise a Each of the one or more microstructuralmodel libraries may simulated T1 time, a simulated T2 time, a simulated proton comprise a number of microstructural models that is within density , a simulated diffusion coefficient , a simulated diffu - a range defined by any two of the preceding values. Different sivity , a simulated fractional anisotropy of diffusion , or a microstructuralmodel libraries may be used to select the one simulated diffusion kurtosis . The simulated MRI parameters or more microstructuralmodels for a given voxel based on may comprise a plurality of simulated MRI parameters . For the region of the brain with which the voxel is associated . instance , the simulated MRI parameters may comprise a The one or more microstructural model libraries may be simulated T1 time and a simulated T2 time, a simulated T1 constructed using the method 200 described herein . time and a simulated diffusion coefficient, a simulated T2 [0094 ] The operation 120 may comprise using the one or time and a simulated diffusion coefficient, or a simulated T1 more computer processors to process the one or more time, a simulated T2 time, and a simulated diffusion coef measured MRIparameters , or a computed function or trans ficient. The number of simulated MRI parameters for each formation of the one or more measured MRI parameters , voxel may be the same across all voxels , all images, or all with the one or more simulated MRI parameters , or a subjects . Alternatively , the number of simulated MRI param computed function or transformation of the one or more eters for each voxel may differ across the voxels , images , or simulated MRI parameters , by computing an objective func subjects . The number of simulated MRIparameters for each tion between the one or more measured MRI parameters , or voxel may be chosen to equal the number ofmeasured MRI a computed function or transformation of the one or more parameters for each voxel. measured MRI parameters , and the one or more simulated 10090 ] The one or more simulated MRI parameters may be MRI parameters , or a computed function or transformation generated from one or more microstructural models at the of the one or more measured MRI parameters , generated voxel. The microstructural models may comprise informa from the one or more microstructural models . The objective tion regarding one or more parameters that may allow function may comprise an Ll norm , an L2 norm , or any computation of one or more of the predicted MRI param other objective function . eters of a voxel described herein . The microstructural mod [0095 ] The objective function may comprise an L1 norm els may comprise information regarding intracellular con computed between a measured MRI parameter and a simu tent of cells that compose brain tissue within a voxel, lated MRI parameter, or an L1 norm computed between any extracellular content of brain tissue within a voxel , a distri 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , or more than 10 measured MRI bution of extracellular content within interstitial space of the parameters and any 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , or more than 10 brain tissue within a voxel, a distribution of intracellular simulated parameters , respectively . For instance , the objec content within intracellular space of cells that compose brain tive function may comprise an Ll norm computed between tissue within a voxel, or brain tissue geometry . a measured T1 time and a simulated T1 time, an L1 norm [0091 ] The microstructural models may comprise mea computed between a measured T2 time and a simulated T2 sured or predicted values of one or more microstructural time, an L1 norm computed between a measured diffusion model parameters such as cell density within a voxel, cell coefficient and a simulated diffusion coefficient, an L1 norm shape within a voxel, cell geometry within a voxel, cell size computed between a measured T1 time and a simulated T1 within a voxel, cell distribution within a voxel, intercellular time, and a measured T2 time and a simulated T2 time, an spacing within a voxel, extracellular matrix homogeneity L1 norm computed between a measured T1 time and a within a voxel , interstitial tortuosity within a voxel , water to simulated T1 time, and a measured diffusion coefficient and protein ratio within a voxel, water to lipid ratio within a a simulated diffusion coefficient, an Ll norm computed voxel, water to carbohydrate ratio within a voxel, protein to between a measured T2 time and a simulated T2 time, and lipid ratio within a voxel, protein to carbohydrate ratio a measured diffusion coefficient and a simulated diffusion within a voxel, or lipid to carbohydrate ratio within a voxel. coefficient , or an Ll norm computed between a measured T1 [ 0092] The one or more microstructural models may be time and a simulated T1 time, a measured T2 time and a informed by knowledge of a region of the brain in which a simulated T2 time, and a measured diffusion coefficient and given voxel is located . For instance, values of the micro a simulated diffusion coefficient. structural model parameters for a voxel associated with a 0096 ] The objective function may comprise an L2 norm particular region of the brain may be assigned based on computed between a measured MRI parameter and a simu experimentally - determined values of the parameters within lated MRI parameter, or an L2 norm computed between any the given region . Alternatively or in combination , values of 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , or more than 10 measured MRI US 2018 /0268942 A1 Sep . 20 , 2018 parameters and any 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , or more than 10 tive or quantitative information related to an extent to which simulated MRI parameters , respectively . For instance , the brain tissue associated with a given voxel has progressed objective function may comprise an L2 norm computed from a healthy state to a diseased state . between a measured T1 time and a simulated T1 time, an L2 [0100 ] The selection of a plurality of diagnostic models norm computed between a measured T2 time and a simu may serve as a check as to whether the method has accu lated T2 time, an L2 norm computed between a measured rately determined the disorder state of the brain tissue diffusion coefficient and a simulated diffusion coefficient , an associated with the voxel . For instance , the plurality of L2 norm computed between a measured T1 time and a diagnostic models may be compared to one another. If most simulated T1 time, and a measured T2 time and a simulated or all of the plurality of diagnostic models are associated T2 time, an L2 norm computed between a measured T1 time with a healthy brain state for a given voxel, this may instill and a simulated T1 time, and a measured diffusion coeffi - greater confidence that the method has accurately deter cient and a simulated diffusion coefficient, an L2 norm mined that brain tissue associated with the voxel is healthy . computed between a measured T2 time and a simulated T2 If most or all of the plurality of diagnostic models are time, and a measured diffusion coefficient and a simulated associated with a diseased brain state for a given voxel, this diffusion coefficient, or an L2 norm computed between a may instill greater confidence that the method has accurately measured T1 time and a simulated T1 time , a measured T2 determined that brain tissue associated with the voxel is time and a simulated T2 time, and a measured diffusion diseased . If the plurality of diagnostic models are not in coefficient and a simulated diffusion coefficient. agreement as to whether the brain tissue is associated with [0097 ] The objective function may comprise a weighted a healthy or a diseased state , this may instill poor confidence L1 norm or a weighted L2 norm . The objective function may that the method has accurately determined the disorder state comprise a Mahalanobis distance . The objective function of brain tissue associated with the voxel. may comprise an explicit formula derived from a single [0101 ] The method 100 may be applied to a single voxel simulated microstructural model or combinations of simu of the plurality of voxels . The method may be applied to lated microstructural models . additional voxels of the plurality of voxels . For instance , [0098 ] In an operation 130 , the method may comprise operations 120 , 130 , and 140 may be repeated one or more selecting one or more diagnostic models from the one or times for additional voxels of the plurality of voxels . The more microstructural models for a voxel of the plurality of method may be applied to all other voxels of the plurality of voxels . The one or more diagnostic models may be selected voxels. For instance , operations 120 , 130 , and 140 may be by computing an objective function for each of the one or repeated for all other voxels of the plurality of voxels . The more microstructural models ; the objective function may be method may be applied to all voxels associated with a any objective function described herein . The objective func specified region of the brain . For instance , operations 120 , tion for each microstructural model may be tested against a 130 , and 140 may be repeated one or more times for all threshold and one or more microstructural models may be voxels associated with a specified region of the brain . The selected as a diagnostic model if the objective function for method may be applied to all voxels associated with an the given microstructural model meets a threshold congru entirety of the brain . For instance , operations 120 , 130 , and ence . In some cases , one diagnostic model may be chosen 140 may be repeated one or more times for all voxels ( such as the diagnostic model which minimizes the objective associated with an entirety of the brain . The method may be function ) . In other cases , a plurality of diagnostic models , applied to a plurality of MRI images . For instance, opera such as at least 2 , at least 3 , at least 4 , at least 5 , at least 6 , tions 110 , 120 , 130 , and 140 may be repeated one or more at least 7 , at least 8 , at least 9 , at least 10 , or at least 10 times for a plurality ofMRI images . Each MRI image of the diagnostic models may be chosen ( for instance , when a plurality of images may be associated with a brain selected plurality of microstructural models meets the threshold from a plurality of brains. Each brain of the plurality of congruence ) . brains may be associated with a subject selected from a [ 0099 ] In an operation 140 , the method may comprise plurality of subjects. using the one or more diagnostic models to determine the [0102 ] FIG . 2 shows a method 200 for constructing a disorder state of the brain tissue associated with a voxel of microstructural model library. the plurality of voxels . The one or more diagnostic models [0103 ] In a first operation 210 , the method may comprise may indicate a healthy brain tissue state, based on knowl creating a first microstructural model corresponding to a edge of the microstructure associated with the diagnostic brain state that is not associated with a disorder. models . For instance , diagnostic models that are similar in [0104 ] In a second operation 220 , the method may com microstructure to a known healthy microstructure may be prise iteratively subjecting the first microstructuralmodel to indicative of a healthy brain tissue state ; alternatively or in a perturbation . Each iteration may produce an additional combination , diagnostic models that are dissimilar in micro - perturbed microstructural model. The first microstructural structure to a known diseased microstructure may be indica model may be selected based on knowledge of the brain tive of a healthy brain tissue state . The one or more diag region associated with the voxel. For instance , the first nostic models may indicate a diseased brain tissue state , microstructural model may be informed by knowledge of a again based on knowledge of the microstructure associated region of the brain in which a given voxel is located , as with the diagnostic models . For instance , diagnostic models described herein . that are dissimilar in microstructure to a known healthy 0105 ]. The first microstructural model may be subjected to microstructure may be indicative of a diseased brain tissue at least 100 iterations, at least 200 iterations, at least 500 state ; alternatively or in combination , diagnostic models that iterations , at least 1 , 000 iterations , at least 2 ,000 iterations, are similar in microstructure to a known diseased micro at least 5 ,000 iterations, at least 10 ,000 iterations, at least structure may be indicative of a diseased brain tissue state . 20 , 000 iterations , at least 50 ,000 iterations , at least 100 ,000 The one or more diagnostic models may comprise qualita iterations, at least 200 , 000 iterations , at least 500 , 000 itera US 2018 /0268942 A1 Sep . 20 , 2018 tions , or at least 1 , 000 , 000 iterations to generate at least 100 [0111 ] Methods of the present disclosure may be used to perturbed microstructural models , at least 200 perturbed determine the disorder state across the brain tissue associ microstructural models , at least 500 perturbed microstruc ated with a plurality of subjects at an accuracy greater than tural models , at least 1 ,000 perturbed microstructural mod or equal to about 60 % , 65 % , 70 % , 75 % , 80 % , 85 % , 90 % , els , at least 2 , 000 perturbed microstructural models , at least 95 % , 99 % , or greater . The methods may be used to deter 5 , 000 perturbed microstructural models , at least 10 ,000 mine the disorder state across the brain tissue associated perturbed microstructural models , at least 20 , 000 perturbed with a plurality of subjects at an accuracy that is within a microstructural models , at least 50 ,000 perturbed micro range defined by any two of the preceding values. structuralmodels , at least 100 ,000 perturbed microstructural [0112 ] Methods of the present disclosure may be used to models , at least 200 , 000 perturbed microstructural models , diagnose a non - neurodegenerative disorder. The non -neuro at least 500 ,000 perturbed microstructuralmodels , or at least degenerative disorder may be a primary neoplasm , a meta 1 ,000 ,000 perturbed microstructural models , respectively . static neoplasm , a seizure disorder, a seizure disorder with The first microstructural model may be subjected to a focal cortical dysplasia, a demyelinating disease ( such as number of iterations that is within a range defined by any multiple sclerosis ) , a non -neurodegenerative encephalopa two of the preceding values to generate a number of per thy (such as a hypertensive encephalopathy, an ischemic turbed microstructuralmodels that is within a range defined encephalopathy, a metabolic encephalopathy, or an infec by any two of the preceding values . tious encephalopathy ) , a cerebrovascular disease ( such as f0106 ] Each iteration may comprise one or more opera stroke or transient ischemic attack ) , or a psychological tions that alter one or more parameters of the first micro disorder ( such as schizophrenia , a schizophreniform disor structural model or subsequent altered iterations of the first der, autism , an autism spectrum disorder , depression , bipolar microstructural model. The one or more operations may disorder, or obsessive compulsive disorder ). comprise depleting cells , altering cellular morphology or [0113 ] Methods of the present disclosure may be used to distribution , altering intracellular or interstitial physico diagnose a neurodegenerative disorder . In some cases, the chemical composition or distribution , altering extracellular methods may be used to diagnose a neurodegenerative matrix composition or distribution , or altering intercellular disorder at least 5 years , at least 10 years , at least 15 years , spacing . Each iteration may comprise a stochastic proce or at least 20 years prior to the development of symptoms dure , such as a Monte Carlo procedure . associated with the neurodegenerative disorder. The meth 0107 ] Many variations , alterations, and adaptations based ods may be used to diagnose a neurodegenerative disorder in on the methods 100 or 200 provided herein are possible . For a time period prior to the development of symptoms asso example , the order of the operations of the methods 100 or ciated with the neurodegenerative disorder that is within a 200 may be changed , some of the operations removed , some range defined by any two of the preceding values . The of the operations duplicated , and additional operations neurodegenerative disorder may be Alzheimer ' s disease , a added as appropriate . Some of the operations may be per non - Alzheimer ' s dementia disorder , Parkinson ' s disease , a formed in succession . Some of the operations may be Parkinsonism disorder , a motor neuron disease (such as performed in parallel . Some of the operations may be amyotrophic lateral sclerosis ) , Huntington ' s disease , a Hun performed once . Some of the operations may be performed tington ' s disease - like syndrome, a transmissible spongiform more than once . Some of the operations may comprise encephalopathy , chronic traumatic encephalopathy, a tauopathy ( such as Pick ' s disease , corticobasal degenera sub - operations. Some of the operations may be automated tion , progressive supranuclear palsy, or Nieman -Pick dis and some of the operations may be manual. ease ) , or any other neurodegenerative disorder. [0108 ] The methods 100 or 200 described herein may be [0114 ] Methods of the present disclosure may further used to determine the disorder state of the brain tissue comprise constructing one or more brain maps . The one or associated with a voxel at any accuracy greater than or equal more brain maps may indicate the neurodegenerative disor to 60 % , 65 % , 70 % , 75 % , 80 % , 85 % , 90 % , 95 % , 99 % , or der state of the brain tissue associated with each voxel of a greater. The methods may be used to determine the disorder plurality of voxels. The methods may comprise display of state of the brain tissue associated with a voxel at an the one or more brain maps on a graphical user interface accuracy that is within a range defined by any two of the (GUI ) of an electronic device of a user. preceding values . [0115 ] The one or more brain maps may comprise a [0109 ] Methods of the present disclosure may be used to qualitative abnormality map ( such as a qualitative neurode determine the disorder state across the brain tissue associ generation map ) . The qualitative abnormality map may ated with a specified region of the brain at an accuracy display whether brain tissue associated with a given voxel greater than or equal to about 60 % , 65 % , 70 % , 75 % , 80 % , displays a microstructure consistent with a brain disorder 85 % , 90 % , 95 % , 99 % , or greater. The methods may be used (such as a neurodegenerative disorder ) , for each voxel of the to determine the disorder state across the brain tissue asso plurality of voxels . The qualitative abnormality map may be ciated with a specified region of the brain at an accuracy that a binary map , with each voxel assigned a microstructure is within a range defined by any two of the preceding values. consistent with a brain disorder displayed in the same color [0110 ] Methods of the present disclosure may be used to ( such as gray or red ) on the qualitative abnormality map . The determine the disorder state across the brain tissue associ determination of whether the given voxel displays a micro ated with the whole brain of a subject at an accuracy greater structure that is consistent with a brain disorder may be than or equal to about 60 % , 65 % , 70 % , 75 % , 80 % , 85 % , subject to a thresholding procedure . For instance , a quali 90 % , 95 % , 99 % , or greater . The methods may be used to tative neurodegeneration map may only indicate that a given determine the disorder state across the brain tissue associ voxel is indicative of a neurodegenerative disorder if the ated with the whole brain of a subject at an accuracy that is microstructure associated with the voxel displays some within a range defined by any two of the preceding values . threshold level ofneurodegeneration . The thresholding pro US 2018 /0268942 A1 Sep . 20 , 2018

cedure may allow a viewer of the qualitative neurodegen - ciated the given brain region displays a microstructure that eration map to ignore minimal neurodegeneration and to is consistent with a brain disorder may be subject to a instead focus their attention on more severely compromised thresholding procedure . The ND map may be a continuous areas of the brain . The qualitative abnormality map may be map , with each brain region assigned a microstructure a percent abnormality map ( such as a percent neurodegen consistent with a neurodegenerative disorder displayed in a eration (PND ) map ) that indicates a percentage of a subject' s color representing the extent to which the brain tissue at the brain ( or region of a subject ' s brain ) that displays tissue given brain region has been damaged by the neurodegen microstructure consistent with a brain disorder (such as a erative disorder on the QND map . For instance , the OND neurodegenerative disorder ). map may display brain regions associated with brain tissue [0116 ] The qualitative abnormality map may indicate that shows no evidence of neurodegeneration displayed in whether brain tissue associated with a given brain region one color (such as blue ) , brain regions associated with brain displays a microstructure consistent with a brain disorder tissue that shows evidence of extensive neurodegeneration ( such as a neurodegenerative disorder ) , for each brain region shown in another color (such as red ) , and brain regions of the plurality of brain regions. The qualitative abnormality associated with brain tissue that shows evidence of inter map may be a binary map , with each brain region assigned mediate neurodegeneration shown in other colors ( such as a microstructure consistent with a brain disorder displayed yellow or orange ) based on the extent to which the brain in the same color ( such as gray or red ) on the qualitative tissue at the given brain region has been damaged by the abnormality map . The determination of whether the given neurodegenerative disorder . Alternatively , the QND map brain region displays a microstructure that is consistent with may use a gradient of a single color to represent the extent a brain disorder may be subject to a thresholding procedure . to which each brain region has been damaged by the For instance, a qualitative neurodegeneration map may only neurodegenerative disorder. The QND map may use a gra indicate that a given brain region is indicative of a neuro dient of a single color ( such as gray ) to represent the extent degenerative disorder if the microstructure associated with of normal variation in the brain regions that have not been the brain region displays some threshold level of neurode damaged by the neurodegenerative disorder. The QND map generation . The thresholding procedure may allow a viewer may use any color scheme. of the qualitative neurodegeneration map to ignore minimal [0120 ] Methods of the present disclosure may further neurodegeneration and to instead focus their attention on comprise constructing one or more data tables. The one or more severely compromised areas of the brain . more data tables may indicate the neurodegenerative disor [0117 ] Alternatively or in combination , the one or more der state of the brain tissue associated with each voxel of a brain maps may comprise a quantitative abnormality map , plurality of voxels . The neurodegenerative disorder state of such as a quantitative neurodegeneration (CND ) map . the brain tissue associated with each voxel may be repre [0118 ] The QND map may display the extent to which sented by a quantitative neurodegeneration ( QND ) score . brain tissue associated with a given voxel displays a micro Alternatively or in combination , the one or more data tables structure consistent with a neurodegenerative disorder, for may indicate the neurodegenerative disorder state of the each voxel of the plurality of voxels . The determination of brain tissue associated with one or more regions of a brain . the extent to which the brain tissue associated the given The neurodegenerative disorder state of the brain tissue voxel displays a microstructure that is consistent with a associated with each region may be represented by a percent brain disorder may be subject to a thresholding procedure . neurodegeneration (PND ) score and / or a quantitative neu The QND map may be a continuous map , with each voxel rodegeneration ( QND ) score and /or any other representative assigned a microstructure consistent with a neurodegenera independent or composite score or scores . The one or more tive disorder displayed in a color representing the extent to data tables may indicate the neurodegenerative disorder state which the brain tissue at the given voxel has been damaged of the entirety of a brain . The neurodegenerative disorder by the neurodegenerative disorder on the QND map . For state of the entirety of a brain may be represented by a instance , the OND map may display voxels associated with percent neurodegeneration (PND ) score and / or quantitative brain tissue that shows little evidence of neurodegeneration neurodegeneration (QND ) score and / or any other represen displayed in one color ( such as blue ) , voxels associated with tative independent or composite score . The PND score may brain tissue that shows evidence of extensive neurodegen indicate a percentage of a subject ' s brain ( or percentage of eration shown in another color (such as red ) , and voxels a region or regions of a subject ' s brain ) that displays tissue associated with brain tissue that shows evidence of inter microstructure consistent with a brain disorder ( such as a mediate neurodegeneration shown in other colors ( such as neurodegenerative disorder ) . The QND score may indicate yellow or orange ) based on the extent to which the brain the extent to which the brain tissue associated with a given tissue at the given voxel has been damaged by the neuro voxel, or a given region of a subject' s brain , or the entirety degenerative disorder . Alternatively , the QND map may use of a subject ' s brain , displays tissue microstructure consistent a gradient of a single color to represent the extent to which with a brain disorder ( such as a neurodegenerative disorder ) . each voxel has been damaged by the neurodegenerative [0121 ] The other representative independent or composite disorder . The QND map may use a gradient of a single color score or scores described herein may comprise a mathemati ( such as gray ) to represent the extent of normal variation in cal combination of multiple measures for a given voxel, the voxels that have not been damaged by the neurodegen plurality of voxels , region , plurality of regions, or whole erative disorder . The QND map may use any color scheme. brains. For example , the composite score may be an esti [ 0119] The QND map may display the extent to which mated neurodegeneration score (END score ) comprising a brain tissue associated with a given brain region displays a mathematical operation of more than one region measure microstructure consistent with a neurodegenerative disorder , such as the product of PND and QND . The mathematical for each brain region of the plurality of brain regions . The operation may comprise multiplying PND and QND scores determination of the extent to which the brain tissue asso and dividing the result by 100 . The other representative US 2018 /0268942 A1 Sep . 20 , 2018 14 independent or composite score or scores may be derived by maceutical. Themethods , alone or in combination with prior computing other parameters such as heterogeneity , asym methods , may also allow more accurate selection of patients metry , or clustering of the voxels or brain regions that for clinical trials . For instance , the methods may ensure that display a microstructure that is consistent with a brain only those subjects displaying certain levels or patterns of disorder , within a brain region , a plurality of brain regions , neurodegeneration are included in a given clinical trial. or an entirety of a subject ' s brain . The determination of [0124 ] Methods of the present disclosure may enable whether a given voxel, a plurality of voxels, a region , a monitoring of brain disorders (such as neurodegenerative plurality of regions of a subject' s brain or a plurality of disorders ) at a plurality of time points, such as at least 2 , at subjects ' brains display a microstructure that is consistent least 5 , at least 10 , at least 20 , at least 50 , at least 100 , at least with a brain disorder may be subject to a thresholding 200 , at least 500 , or at least 1 , 000 time points . The methods procedure . The other representative independent or compos may enable monitoring of brain disorders for a number of ite score or scores may be derived by computing other time points that is within a range defined by any two of the parameters such as heterogeneity , asymmetry , or clustering preceding values . Each pair of the plurality of time points of the voxels or brain regions that display a microstructure may be separated by a plurality of time intervals . For that is consistent with a healthy state of brain tissue . The instance , each pair of time points may be separated by at other representative independent or composite score or least 1 day , at least 2 days, at least 5 days, at least 1 week , scores may be a whole brain score (WBS ) . The WBS may at least 2 weeks, at least 1 month , at least 2 months , at least indicate the extent to which the entirety of a subject' s brain 5 months, at least 10 months , at least 1 year, at least 2 years , displays tissue microstructure consistent with a brain disor at least 5 years , at least 10 years, at least 20 years , at least der ( such as a neurodegenerative disorder ). The whole brain 50 years , or at least 100 years . Each of the time points may score may be expressed by a real number. Similarly , the be separated by a period of time that is within a range PND score , the QND score , and the other representative defined by any two of the preceding values . In this manner, independent or composite score or scores may be expressed the methods may be used to track the development or by a real number. progression of a brain disorder over a period of time. [0122 ] Data produced by methods of the present disclo [0125 ] Though described herein with respect to determin sure , alone or in combination with the data tables described ing a disorder state of brain tissue , the methods and systems herein or the brain maps described herein , may be analyzed of the present disclosure , such as the methods 100 and 200 , using machine learning procedures to improve the accuracy may be utilized to determine a state ( e . g ., disorder state ) of of diagnosis of neurodegenerative disorders . The machine other tissues . For instance , methods and systems of the learning procedures may comprise various supervised present disclosure may be utilized to determine a disorder machine learning techniques , various semi- supervised state of spinal cord tissue, heart tissue, vascular tissue, lung machine learning techniques , and / or various unsupervised tissue, liver tissue, kidney tissue , esophageal tissue, stomach machine learning techniques . For instance , the machine tissue , intestinal tissue, pancreatic tissue , thyroid tissue , learning procedures may utilize autoencoders , stacked auto adrenal tissue , spleen tissue, lymphatic tissue, appendix encoders, neural networks , convolutional neural networks , tissue, breast tissue , bladder tissue , vaginal tissue , ovarian alternating decision trees ( ADTree ) , Decision Stumps, func tissue, uterine tissue, penile tissue, testicular tissue , prostatic tional trees (FT ), logistic model trees (LMT ) , logistic regres tissue , skeletal muscle tissue, skin , or non -brain tissues of sion , Random Forests , linear classifiers , factor analysis , the head and neck (such as soft tissues of the skull base , principle component analysis , neighborhood component tissues of facial structures such as the eyes, nose or ears , analysis , sparse filtering , stochastic neighbor embedding , or tissues of the oral cavity such as the tongue , uvula , gingiva , any other machine learning algorithm or statistical algo or palatine tonsils , or deep structures of the neck such as the rithm . One or more algorithms may be used together to retropharyngeal space , the para -pharyngeal space , epiglottis , generate an ensemble method , wherein the ensemble method larynx , or trachea ). may be optimized using a machine learning ensemble meta [0126 ] Though described herein with respect to analysis of algorithm such as a boosting ( e . g . , AdaBoost, LPBoost , MRI images , the methods and systems of the present dis TotalBoost , BrownBoost , MadaBoost, LogitBoost , etc . ) to closure , such as the methods 100 and 200 , may be utilized reduce bias and /or variance . Machine learning analyses may to analyze images obtained by other medical imaging tech be performed using one or more of various programming nologies . For instance , the methods may allow analysis of languages and platforms, such as R , Weka , Python , and /or images obtained through X - ray computed tomography (CT ) Matlab , for example . Machine learning analyses may be imaging , single photon emission computed tomography performed using a machine learning platform , such as (SPECT ) imaging , electron paramagnetic resonance (EPR ) BigML . imaging , positron emission tomography (PET ) imaging , [ 0123] Methods of the present disclosure may be used to ultrasound imaging , or any combination of such imaging inform drug development. For instance , the methodsmay be technologies . used to assess the efficacy of pharmaceutical interventions [0127 ] FIG . 3 shows a system 300 for determining a for neurodegenerative disorders . Since the methods may disorder state of brain tissue in a brain of a subject. The allow diagnosis of a neurodegenerative disorder during the system may comprise a database 310 . The database may earliest stages of the disorder , the methods may allow comprise any MRI data described herein . For instance , the pharmaceuticals to be tested on a cohort of subjects at a database may comprise any MRI data described herein with much earlier stage in the progression of the neurodegenera respect to the method 100 or the method 200 . The system tive disorder, when minimal damage to brain tissue has may further comprise one or more computer processors 320 . occurred and pharmaceutical interventions may be more The one or more processors may be individually or collec effective . The methods may allow accurate tracking of tively programmed to implement any of the methods neurodegeneration following the administration of a phar described herein . For instance, the one or more processors US 2018 /0268942 A1 Sep . 20, 2018 may be individually or collectively programmed to imple 430 . For instance , the computer system 401 can communi ment any or all operations of the methods of the present cate with a remote computer system of a user . Examples of disclosure , such as methods 100 or 200 . remote computer systems include personal computers ( e . g . , [0128 ] FIG . 4 shows a computer system 401 that is pro portable PC ), slate or tablet PC ' s ( e . g . , Apple® iPad , Sam grammed or otherwise configured to operate a system or sung® Galaxy Tab ) , telephones , Smart phones ( e . g ., Apple® method for determining a disorder state of brain tissue in a iPhone , Android - enabled device , Blackberry® ) , or personal subject described herein . The computer system 401 can digital assistants . The user can access the computer system regulate various aspects of the present disclosure . The 401 via the network 430 . computer system 401 can be an electronic device of a user [0134 ] Methods as described herein can be implemented or a computer system that is remotely located with respect by way of machine ( e. g ., computer processor) executable to the electronic device. The electronic device can be a code stored on an electronic storage location of the computer stationary electronic device such as a desktop computer. The system 401 , such as , for example , on the memory 410 or electronic device can be a mobile electronic device. electronic storage unit 415 . The machine executable or [ 0129 ] The computer system 401 includes a central pro machine readable code can be provided in the form of cessing unit (CPU , also “ processor” and “ computer proces software . During use , the code can be executed by the sor” herein ) 405 , which can be a single core or multi core processor 405 . In some cases, the code can be retrieved from processor, or a plurality of processors for parallel process the storage unit 415 and stored on the memory 410 for ready ing . The computer system 401 also includes memory or access by the processor 405 . In some situations, the elec memory location 410 ( e . g ., random - access memory , read tronic storage unit 415 can be precluded , and machine only memory, flash memory ) , electronic storage unit 415 executable instructions are stored on memory 410 . ( e . g . , hard disk ) , communication interface 420 ( e . g ., network [0135 ] The code can be pre - compiled and configured for adapter ) for communicating with one ormore other systems, use with a machine having a processer adapted to execute and peripheral devices 425 , such as cache , other memory , the code, or can be compiled during runtime. The code can data storage and /or electronic display adapters. The memory be supplied in a programming language that can be selected 410 , storage unit 415 , interface 420 and peripheral devices to enable the code to execute in a pre - compiled or as 425 are in communication with the CPU 405 through a compiled fashion . communication bus ( solid lines ) , such as a motherboard . The [0136 ] Aspects of the systems and methods provided storage unit 415 can be a data storage unit (or data reposi herein , such as the computer system 401 , can be embodied tory ) for storing data . The computer system 401 can be in programming Various aspects of the technology may be operatively coupled to a computer network (" network ” ) 430 thought of as “ products ” or “ articles of manufacture” typi with the aid of the communication interface 420 . The cally in the form of machine ( or processor) executable code network 430 can be the Internet, an internet and /or extranet , and /or associated data that is carried on or embodied in a or an intranet and / or extranet that is in communication with type of machine readable medium . Machine - executable the Internet. The network 430 in some cases is a telecom code can be stored on an electronic storage unit, such as munication and / or data network . The network 430 can memory (e .g ., read -only memory , random -access memory, include one or more computer servers , which can enable flash memory ) or a hard disk . “ Storage ” type media can distributed computing , such as cloud computing . The net include any or all of the tangible memory of the computers , work 430 , in some cases with the aid of the computer system processors or the like, or associated modules thereof, such as 401, can implement a peer - to - peer network , which may various semiconductor memories , tape drives , disk drives enable devices coupled to the computer system 401 to and the like , which may provide non - transitory storage at behave as a client or a server . any time for the software programming . All or portions of [ 0130 The CPU 405 can execute a sequence of machine the software may at times be communicated through the readable instructions, which can be embodied in a program Internet or various other telecommunication networks. Such or software . The instructions may be stored in a memory communications , for example , may enable loading of the location , such as the memory 410 . The instructions can be software from one computer or processor into another, for directed to the CPU 405 , which can subsequently program example , from a management server or host computer into or otherwise configure the CPU 405 to implementmethods the computer platform of an application server . Thus , of the present disclosure . Examples of operations performed another type of media that may bear the software elements by the CPU 405 can include fetch , decode, execute , and includes optical, electrical and electromagnetic waves , such writeback . as used across physical interfaces between local devices , [0131 ] The CPU 405 can be part of a circuit, such as an through wired and optical landline networks and over vari integrated circuit . One or more other components of the ous air - links . The physical elements that carry such waves , system 401 can be included in the circuit . In some cases , the such as wired or wireless links, optical links or the like, also circuit is an application specific integrated circuit (ASIC ) . may be considered as media bearing the software . As used [0132 ] The storage unit 415 can store files, such as drivers , herein , unless restricted to non -transitory , tangible “ storage ” libraries and saved programs. The storage unit 415 can store media , terms such as computer or “ machine readable user data , e . g ., user preferences and user programs. The medium ” refer to any medium that participates in providing computer system 401 in some cases can include one or more instructions to a processor for execution . additional data storage units that are external to the com [0137 ] Hence , a machine readable medium , such as com puter system 401 , such as located on a remote server that is puter -executable code , may take many forms, including but in communication with the computer system 401 through an not limited to , a tangible storage medium , a carrier wave intranet or the Internet . medium or physical transmission medium . Non - volatile [ 0133 ] The computer system 401 can communicate with storage media include , for example , optical or magnetic one ormore remote computer systems through the network disks , such as any of the storage devices in any computer ( s ) US 2018 /0268942 A1 Sep . 20 , 2018 or the like , such as may be used to implement the databases , model includes known measured or predicted values of cell etc . shown in the drawings . Volatile storage media include density , cell shape , cell geometry , cell size, intercellular dynamic memory , such as main memory of such a computer spacing , extracellular matrix heterogeneity , interstitial tor platform . Tangible transmission media include coaxial tuosity , water to lipid ratio , and other tissue parameters that cables ; copper wire and fiber optics , including the wires that can influence structural and diffusion measurements . Typical comprise a bus within a computer system . Carrier - wave ensembles consist of 1024x1024x1024 cell arrays. FIG . 5A transmission media may take the form of electric or elec shows a portion of a cell array for a normal tissue micro tromagnetic signals , or acoustic or light waves such as those structural model. As shown in FIG . 5A , gray cubes 510 generated during radio frequency ( RF ) and infrared ( IR ) data represent brain cells and green spots 520 represent molecu communications. Common forms of computer -readable lar obstacles of the extracellular space . media therefore include for example : a floppy disk , a flexible 0143 ] The normal tissue microstructural model is then disk , hard disk , magnetic tape , any other magnetic medium , used for finite -element and Monte Carlo simulations of the a CD -ROM , DVD or DVD -ROM , any other optical tissue chemical composition , tissue micro - lattice topogra medium , punch cards, paper tape, any other physical storage phy, and molecular kinetics , which can subsequently be used medium with patterns of holes , a RAM , a ROM , a PROM to generate predicted structural MRI signals ( T1 -weighted , and EPROM , a FLASH -EPROM , any other memory chip or T2 -weighted , and diffusion -weighted MRI signals ) and the cartridge, a carrier wave transporting data or instructions, associated bulk diffusion coefficients . By modifying the cables or links transporting such a carrier wave , or any other range of input parameters in the model , the sensitivity of the medium from which a computer may read programming output signal values is determined . The reconstructed tissue code and/ or data . Many of these forms of computer readable and calculated values can be directly correlated to MRI media may be involved in carrying one or more sequences values at a single voxel. FIG . 5B shows a sample Monte of one or more instructions to a processor for execution . Carlo simulation with delivery of freely moving molecules . (0138 ] The computer system 401 can include or be in As shown in FIG . 5B , red dots represent areas of abnormal communication with an electronic display 435 that com molecular distribution and blue dots represent areas of prises a user interface (UI ) 440 . Examples of UI' s include , normal molecular distribution . without limitation , a graphical user interface (GUI ) and [0144 ] By manipulating structural components indepen web -based user interface . dently and in concert (such as depleting cells , altering [ 0139 ] Methods and systems of the present disclosure can morphology , altering interstitial obstructions, etc . ) accord be implemented by way of one or more algorithms. An ing to reported and predicted variations in brain tissue , the algorithm can be implemented by way of software upon platform may be used to generate a continuous range of execution by the central processing unit 405 . The algorithm tissue transitions from healthy to severe degeneration . FIG . can , for example , determine a disorder state of brain tissue 5C shows representative models ofhealthy and degenerating in a subject described herein . brain tissue. 10140 ) While preferred embodiments of the present inven [0145 ] Through in - silico modeling, a database consisting tion have been shown and described herein , it will be of many possible variations of tissue structure that represent obvious to those skilled in the art that such embodiments are a range of healthy and diseased states that can be directly provided by way of example only . It is not intended that the translated to MRI scan values has been assembled . With this invention be limited by the specific examples provided reference set correlating a variation of microstructure com within the specification . While the invention has been position to MR signals , real human T1 -weighted , described with reference to the aforementioned specifica T2 -weighted and diffusion -weighted MRI scans can be tion , the descriptions and illustrations of the embodiments applied to predict the most probable tissue microstructure herein are not meant to be construed in a limiting sense . contained in each MRI voxel . The platform may thus be Numerous variations , changes, and substitutions will now occur to those skilled in the art without departing from the regarded as providing a virtual tissue microscope . invention . Furthermore, it shall be understood that all aspects of the invention are not limited to the specific Example 2: Interpretation of Human MRI depictions, configurations or relative proportions set forth [0146 ] FIG . 6 shows the processing of human MRI images herein which depend upon a variety of conditions and to produce neurodegeneration maps . variables . It should be understood that various alternatives to 101471 With a rigorous pipeline for structural prediction of the embodiments of the invention described herein may be MRI, each voxel of the human brain scan can be compre employed in practicing the invention . It is therefore con hensively characterized . The resulting output is a gray map templated that the invention shall also cover any such with voxel -wise intensity scaling to represent the predicted alternatives , modifications , variations or equivalents . It is deviation from normal. intended that the following claims define the scope of the [0148 ] Based on literature knowledge and initial experi invention and thatmethods and structures within the scope mental measurements , it is determined if the calculated of these claims and their equivalents be covered thereby . tissue structure from a given combination of T1 time, T2 time, and diffusion coefficient values is within tolerance for EXAMPLES healthy brain tissue or resides in the spectrum of abnormal ity . Each voxel 610 determined abnormal is coded red or a Example 1: Relation of Microstructure to MRI variable color, respectively denoting binary ( percent neuro [ 0141 ] FIG . 5 shows the construction of exemplary micro degeneration , PND ) or quantitative ( quantitative neurode structural models . generation , QND ) abnormality in output maps. In effect , 0142 ] The first step is construction of a normal tissue PND identifies the abnormal voxels ( red ) and OND defines microstructural model. The normal tissue microstructural how abnormal those voxels are . The range is from dark blue US 2018 /0268942 A1 Sep . 20 , 2018 17

620 ( close to normal) to light blue 630 ( slightly abnormal) based on calculated PND and roughness in the female to yellow 640 (moderately abnormal) to dark red 650 (very patient scans. These raw data are suggestive of an ability to abnormal) . diagnose some individuals with MCI. The bigger implica tions, however, are that systems and methods for analyzing Example 3: Generation of PND and QND Brain brain MRI are providing meaningful estimates of early Maps changes in microstructural state associated with disease . [0149 ] FIG . 7 shows exemplary MRI, PND , and QND Example 5 : Diagnosis by Machine Learning brain maps from a diseased individual. The MRI scan shown in FIG . 7 is a diffusion -weighted image that, combined with 10154 ] With an extensive list of input parameters and T1 - weighted and T2 - weighted images (not shown in FIG . output calculations that vary across an anatomically variable 7 ) , is used to generate a grayscale output image . The image stack , the results can be dramatically improved by grayscale output image serves as the underlying image for applying iterative machine learning on a subset of images . the PND and QND maps following removal of the skull and Through machine learning , weighting of known risk factors cerebrospinal fluid domains . Red voxels in the PND map such as age and gender and less obvious regional patterns of represent microstructures that deviate from predicted ranges microstructure estimates, the distinguishing power of the of structural tolerance for normal tissue. Colored voxels in degeneration maps can be significantly improved . To vali the QND map similarly represent abnormal voxels , with date the diagnostic capacity, the blind ADNI analysis was color added to code for the extent of abnormality (blue = low processed through a BigML algorithm and a predicted abnormality , red = high abnormality ) . diagnosis was assigned to each image set . The resulting data [0150 ] To assess the degenerative state of an individual are presented as a confusion matrix in Table 1 . scan , the abnormality relative coverage ( PND ) , extent of degeneration (QND ), estimated total degeneration (END ) , TABLE 1 variation of gray value ( gRou ), and variation of color value (color roughness , cRou ) were further characterized . Confusion matrix [ 0151] FIG . 8 shows exemplary brain maps for a young , Clinical Diagnosis asymptomatic brain , a normal aged brain with no detected neurodegenerative symptoms, and an aged brain with clini Disease No disease cal symptoms of severe degeneration . The grayscale maps Diagnosis Disease 643 34 0 . 95 reflect determined microstructure at each voxel of each by BigML No disease 16 172 0 . 91 two - dimensional MRI image . 0 .98 0 .83 Example 4 : Assessment of a Cohort of Subjects [0155 ] The sensitivity or ability to positively detect dis eased patients using BigML for categorical prediction ( de 10152] To assess the accuracy of the systems and methods fined here as clinically diagnosed with either MCI or AD ) for detecting neurodegeneration described herein , a rigorous was 98 % . The specificity was 93 % . The positive predictive microstructure prediction analysis was performed on avail value was 95 % . The negative predictive value was 91 % . able MRI data in the Alzheimer 's Disease Neuroimaging Overall accuracy was slightly lower, at 90 % , owing to the Initiative (ADNI ) database . Scans from healthy individuals and a range of degenerative states, including early and late predicted abnormality in clinically undiagnosed individuals . cognitive impairment and diagnosed AD , were blindly pro Whereas some of these individuals may not progress toward cessed using the systems and methods described herein and a clinical diagnosis , there is a possibility that the systems statistics were generated for each voxel of each image slice . and methods described herein are detecting structural Brain outputmaps were normalized to 50 slices for approxi changes before any cognitive symptoms are present. This mate region registration . Mean values were calculated for population bordering between normal and abnormal is the groups segregated by gender, clinical diagnosis , and output most difficult yet arguably the most important to address . parameter ( e. g ., PND , QND , END , gRou , and cRou shown in FIG . 9 ) . Each essential evaluation parameter was deter Example 6 : Longitudinal Brain Imaging Studies mined across normalized brain regions and with gender [0156 ] To evaluate the ability to predict early onset of segregation . The population was as follows: 152 individuals degenerative disease , a collection of MRI scans were diagnosed with AD ( 98 male , 54 female ), 507 individuals acquired from collaborators performing longitudinal brain showing mild cognitive impairment (MCI ) (317 male , 190 imaging studies. All patients were imaged at the time of MCI female ) , and 206 normal individuals showing no symptoms or AD diagnosis (grouped here as Abnormal, though most of cognitive impairment ( 102 male , 104 female ) ] . patients were diagnosed with severe MCI ) , and most [ 0153] FIG . 9 shows mean plots of degenerative analysis received numerous scans prior to symptom presence and output parameters for the ADNI images . As shown in the clinical diagnosis . Scans were acquired up to 16 years prior mean plots , there is a clear distinction of patients diagnosed to diagnosis . The cohort distribution and time of scans , with AD within the PND and roughness values across the relative to diagnosis date, is shown as a bar plot in FIG . 10 . scan slices , suggesting a robust use in AD diagnosis confir Included are a large number of scans from healthy (Normal ) mation . As expected from a mix of early and late impairment individuals with repeated scans over time in this case , time stages , the MCI values are shifted toward those of the O is themost recent scan and ' years before diagnosis ' is years normal individuals. Considering the estimate that MCI before most recent scan ) to reflect the normal individual patients progress at a modest rate ( 10 - 15 % annually ) toward variation over time. full AD , it is striking that the systems and methods described [0157 ] FIG . 10 shows a population distribution for a herein can discriminate between MCI and normal patients longitudinal study to evaluate early detection of Alzheimer 's US 2018 /0268942 A1 Sep . 20 , 2018 18 disease. Each of the Abnormal patients was diagnosed with Example 8 : Region -by - Region Diagnosis either MCI or AD . A majority of abnormal patients were diagnosed with late stage MCI. The scans prior to diagnosis [0163 ] FIG . 14 shows optimized single region prediction were pooled into 4 year intervals . Segregation by gender accuracy of diagnosis within the ADNI dataset for a variety shows the population sizes used for the microstructural of brain regions using the systems and methods described degeneration analysis . For the Normal individuals , the time herein . Each set of model parameters is tested for optimal point reflects the time before a most recent scan date . prediction accuracy for each independent brain region . [0158 ] Plots of the principal output values from the degen [0164 ] FIG . 15 shows PND measurement distributions eration maps show markedly higher PND , END and cRou at across subjects of a variety of ages for the whole brain , the time of diagnosis , as shown in FIG . 11 . Most impor cerebellum , thalamus, posterior cingulate , precuneus, and tantly , the raw output values show impressive differentiation hippocampus. of male and female patients relative to normal controls 10165 ] FIG . 16 shows PND measurement distributions averages , even in the earliest scans . The female cohort across subjects of a variety of ages for the entorhinal cortex , shows increased microstructure abnormality at the first time basal ganglia , parietal lobe, occipital lobe , prefrontal cortex , interval before diagnosis , but most of the differentiation is and premotor cortex . lost at earlier time points , possibly due to a limited sample [0166 ] FIG . 17 shows PND measurement distributions size or an increasingly lower amounts of, or total absence of, across subjects of a variety of ages for the precentral gyrus, microstructural abnormality in very early time points . postcentral gyrus , temporal lobe , paracentral lobule, olfac Remarkably , the male population PND , QND and END tory bulb , and anterior -mid cingulum . remain elevated above average control values through even the earliest time points . This provides evidence for detection Example 8 : Machine Learning Prediction from of early degeneration in patients before irreversible degen Optimized ADNI Image Processing eration occurs and many years before current practices effectively diagnose the changes . [0167 ] FIG . 18 shows attainable AD diagnostic metrics 10159 ] FIG . 11 shows an early detection analysis of MRI using machine learning. Bootstrap - aggregated decision trees images collected longitudinally from the cohort distribution . were applied to predict hierarchical classifiers from opti Output parameters were blindly generated through micro mally processed ADNI images. structure prediction from input MRI scans at each time point. Bars represent mean values and standard deviations Example 9 : Whole Brain Scoring for the respective output measure . [0168 ] FIG . 19 shows a distribution of whole brain scores [0160 ] To confirm the early detection ability of our micro (WBS ) for ADNI subject scans. The score for each brain was structure analysis , output evaluation through BigML pro plotted vs the subject' s age at the time of scan acquisition . cessing was repeated using different parameters . After blind In this example, WBSs were generated through logistic processing , each scan was characterized as normal or abnor regression analysis of regional PND , OND and END values . mal and compared to the known clinical diagnoses . Among The WBS alone provides statistically separated distributions the 40 diagnosed patients ( included here are individuals from the image set without scans prior to diagnosis ) , a of normal and AD -diagnosed individuals . majority was determined to be abnormal, as shown in FIG . 12 . From this extended cohort , nearly 93 % sensitivity was FURTHER ASPECTS OF THE DISCLOSURE achieved at the time of diagnosis. Remarkably, a similar [0169 ] 1. A method for determining a disorder state of 93 % sensitivity was achieved throughout all early scans, brain tissue in a brain of a subject, comprising : including multiple scans collected more than a decade 101701 ( a ) obtaining magnetic resonance imaging before diagnosis , all of which were predicted to be abnor (MRI ) data comprising at least one MRI image of the mal. Though further detailed evaluation and analysis of brain , the MRI image comprising a plurality of voxels , newly collected images will be necessary to verify the a voxel of the plurality of voxels being associated with robustness of our detection system , the initial performance the brain tissue of the brain of the subject and com has surpassed current expectations that exist throughout the prising one or more measured MRI parameters in the field of clinicalbrain imaging . Presented herein is a tool with MRI data ; encouraging performance in detection of early changes in [0171 ] ( b ) for the voxel of the plurality of voxels , using brain tissue structure that can serve as a predictor for future one or more computer processors to process the one or development of degenerative disease . This is a tool desper more measured MRI parameters with one or more ately needed in the medical community and pharmaceutical simulated MRI parameters for the voxel, the one or industry to aid the detection and prevention of Alzheimer ' s more simulated MRI parameters being generated from disease and similar neurodegenerative diseases . one or more microstructural models at the voxel; 10161] FIG . 12 shows a determination of abnormality in a [0172 ] (c ) for the voxel of the plurality of voxels , mixed cohort longitudinal study . All individuals were diag selecting a diagnostic model from the one or more nosed with abnormalities of MCI or AD at time 0 . Scans microstructuralmodels , the diagnostic modelmeeting a prior to diagnosis were grouped in three - year intervals . threshold congruence between the one or more mea sured MRI parameters and the one or more simulated Example 7 : Registration of Brain Images MRI parameters associated with the diagnostic model; [0162 ] FIG . 13 shows the registration or alignment of and subject images to an annotated human brain parcellation [0173 ] ( d ) for the voxel of the plurality of voxels , using atlas . Each row is a subsample of images throughout the the diagnostic model to determine the disorder state of brain from a different imaging axis or orientation . the brain tissue associated with the voxel. US 2018 /0268942 A1 Sep . 20 , 2018 19

[0174 ] 2. The method of aspect 1, wherein each voxel [0189 ] 17 . The method of aspect 15 or 16 , wherein the comprises a plurality of measured MRI parameters. microstructural model library is constructed by : 10190 ] ( a ) creating a first microstructural model corre [0175 ] 3 . The method of aspect 1 or 2 , wherein the one or sponding to a brain state that is not associated with a more measured MRI parameters are a plurality ofmeasured disorder; and MRI parameters . [0191 ] ( b ) iteratively subjecting the first microstructural [ 0176 ] 4 . The method of any one of aspects 1 - 3 , wherein model to a perturbation , each iteration producing an the one ormore simulated MRI parameters are a plurality of additional perturbed microstructural model. simulated MRI parameters . [0192 ] 18 . The method of aspect 17 , wherein ( b ) com [ 0177 ] 5 . The method of any one of aspects 1 - 4, further prises subjecting the first microstructural model to at least 100 iterations to generate at least 100 perturbed microstruc comprising repeating (b )- (d ) one or more times for addi tural models . tional voxels of the plurality of voxels . [0193 ] 19 . The method of aspect 17 or 18 , wherein the first [ 0178 ] 6 . The method of aspect 5 , further comprising microstructural model is selected based on knowledge of the repeating ( b ) - ( d ) for all other voxels of the plurality of brain region associated with the voxel . voxels . [0194 ] 20 . The method of any one of aspects 17 - 19 , [0179 ] 7 . The method of aspect 5 , further comprising wherein the perturbation comprises an operation selected repeating ( b ) - ( d ) for all voxels associated with a specified from the group consisting of: depleting cells , altering cel region of the brain . lular morphology or distribution , altering intracellular or interstitial physico -chemical composition or distribution , [0180 ] 8 . The method of aspect 5 , further comprising altering extracellular matrix composition or distribution , and repeating (b ) - (d ) for all voxels associated with an entirety of altering intercellular spacing . the brain . [0195 ] 21. The method of any one of aspects 17 - 20 , [0181 ] 9 . The method of aspect 5 , further comprising wherein the perturbation comprises a stochastic procedure . repeating ( a ) - ( d ) for a plurality of MRI images, each MRI [01961 22 . The method of any one of aspects 1 - 21 , wherein image of the plurality ofMRI images associated with a brain the threshold congruence is determined by computing an selected from a plurality of brains, each brain of the plurality objective function between the one or more measured MRI of brains associated with a subject selected from a plurality parameters and the one or more simulated MRI parameters. of subjects. [0197 ] 23 . The method of aspect 22 , wherein the objective [0182 ] 10 . The method of any one of aspects 1 - 9 , wherein function comprises an L1 norm or an L2 norm . the MRI image is selected from the group consisting of: a [0198 ] 24 . The method of any one of aspects 1 -23 , wherein longitudinal relaxation time ( T1 )- weighted MRI image , a determining the disorder state of the brain tissue associated with the voxel is achieved at an accuracy of at least 90 % . transverse relaxation time ( T2 ) -weighted MRI image , and a [ 0199 ) 25 . The method of any one of aspects 7 - 24 , wherein diffusion -weighted MRI image . determining the disorder state across the brain tissue asso [0183 ] 11. Themethod of any one of aspects 1 - 10 , wherein ciated with the specified region of the brain is achieved at an the measured MRI parameter is selected from the group accuracy of at least 90 % . consisting of: a longitudinal relaxation time ( T1 ) , a trans [0200 ] 26 . The method of any one of aspects 8 -25 , wherein verse relaxation time ( T2 ), and a diffusion coefficient. determining the disorder state of the brain tissue associated [ 01841 12 . Themethod of any one of aspects 1 - 11 , wherein with the whole brain of the subject is achieved at an the simulated MRI parameter is selected from the group accuracy of at least. consisting of: a longitudinal relaxation time ( T1) , a trans [0201 ] 27 . The method of any one of aspects 9 - 26 , wherein verse relaxation time ( T2 ) , and a diffusion coefficient. determining the disorder state of the brain tissue associated [0185 ] 13 . The method of any one of aspects 1 - 12 , wherein the plurality of subjects is achieved at an accuracy of at least the one or more microstructural models comprise informa 90 % . tion regarding a parameter selected from the group consist [0202 ] 28 . The method of any one of aspects 1 -27 , wherein ing of: intracellular content, extracellular content, distribu the disorder is a non -neurodegenerative disorder. tion of extracellular content within interstitial space , [0203 ] 29 . The method of aspect 28 , wherein the disorder distribution of intracellular content within intracellular is selected from the group consisting of: a primary neo space, and tissue geometry . plasm , a metastatic neoplasm , a seizure disorder, a seizure [0186 ] 14 . The method of any one or aspects 1 -13 , wherein disorder with focal cortical dysplasia , a demyelinating dis the one or more microstructuralmodels comprise measured order, a non -neurodegenerative encephalopathy, a cerebro or predicted values of a parameter selected from the group vascular disease, and a psychological disorder. consisting of: cell density , cell shape , cell geometry , cell [0204 ] 30 . Themethod of any one of aspects 1 - 27 , wherein size, cell distribution , intercellular spacing, extracellular the disorder is a neurodegenerative disorder. matrix homogeneity , interstitial tortuosity , water to protein [0205 ) 31 . The method of aspect 30 , wherein the method enables diagnosis of a neurodegenerative disorder more than ratio , water to lipid ratio , water to carbohydrate ratio , protein 5 years prior to the development of symptoms associated to lipid ratio , protein to carbohydrate ratio , and lipid to with the neurodegenerative disorder. carbohydrate ratio . [0206 ] 32 . The method of aspect 30 or 31, wherein the [0187 ] 15. The method of any one of aspects 1 - 14 , wherein method enables monitoring of the neurodegenerative disor the one or more microstructural models are selected from a der at a plurality of time points , the plurality of time points microstructural model library . separated by a plurality of time intervals . [0188 ] 16 . The method of aspect 15 , wherein the micro [0207 ] 33 . The method of any one of aspects 30 -32 , structural model library comprises at least 100 microstruc wherein the neurodegenerative disorder is selected from the tural models . group consisting of: Alzheimer 's disease , a non -Alzheimer 's US 2018 /0268942 A1 Sep . 20 , 2018 20 dementia disorder, Parkinson 's disease , a Parkinsonism dis the brain tissue of the brain of the subject and com order, a motor neuron disease , Huntington ' s disease , a prising one or more measured MRI parameters in the Huntington ' s disease - like syndrome, transmissible spongi MRI data ; form encephalopathy, chronic traumatic encephalopathy, [0222 ] ( b ) for the voxel of the plurality of voxels , using and a tauopathy. one or more computer processors to process the one or [0208 ] 34 . The method of any one of aspects 1 -33 , further more measured MRI parameters with one or more comprising constructing a brain map that, for each voxel of simulated MRI parameters for the voxel, the one or the plurality of voxels, indicates the disorder state of the more simulated MRI parameters being generated from brain tissue associated with the voxel . one or more microstructural models at the voxel ; [0209 ] 35 . The method of aspect 34 , further comprising [0223 ] ( c ) for the voxel of the plurality of voxels , displaying the brain map on a graphical user interface of an selecting a diagnostic model from the one or more microstructuralmodels , the diagnostic modelmeeting a electronic device of a user . threshold congruence between the one or more mea [ 0210 ] 36 . The method of aspect 34 or 35 , wherein the sured MRI parameters and the one or more simulated brain map comprises a qualitative abnormality map . MRI parameters associated with the diagnostic model; [ 0211 ] 37 . The method of aspect 34 or 35, wherein the and brain map comprises a binary abnormality map . [ 0224 ] (d ) for the voxel of the plurality of voxels , using [0212 ] 38 . The method of aspect 34 or 35 , wherein the the diagnostic model to determine the disorder state of brain map comprises a quantitative abnormality map . the brain tissue associated with the voxel. [0213 ] 39. The method of aspect 34 or 35 , wherein the (02251 43 . The non - transitory computer -readable medium brain map comprises a percent" anomaliy abnormality mapmap . of aspect 42 , wherein each voxel comprises a plurality of [ 0214 ] 40 . A method for determining a disorder state of a measured MRI parameters. tissue in a portion of a body of a subject, comprising : [0226 ] 44 . The non - transitory computer -readable medium [0215 ] ( a ) obtaining magnetic resonance imaging of aspect 42 or 43 , wherein the one or more measured MRI (MRI ) data comprising at least one MRI image of the parameters are a plurality of measured MRI parameters . tissue , the MRI image comprising a plurality of voxels , (0227 ] 45 . The non - transitory computer -readable medium a voxel of the plurality of voxels being associated with of any one of aspects 42 - 44 , wherein the one or more the tissue of the subject and comprising one or more simulated MRI parameters are a plurality of simulated MRI measured MRI parameters in the MRI data ; parameters . [0216 ] (b ) for the voxel of the plurality of voxels , using [0228 ] 46 . The non - transitory computer -readable medium one or more computer processors to process the one or of any one of aspects 42 - 45 , wherein the method further more measured MRI parameters with one or more comprises repeating ( b ) - ( d ) one or more times for additional simulated MRI parameters for the voxel, the one or voxels of the plurality of voxels . more simulated MRI parameters being generated from [0229 ] 47 . The non - transitory computer- readable medium of aspect 46 , wherein the method further comprises repeat one or more microstructural models at the voxel; ing (b ) - ( d ) for all other voxels of the plurality of voxels . [0217 ] ( c ) for the voxel of the plurality of voxels , [02301 48 . The non - transitory computer - readable medium selecting a diagnostic model from the one or more of aspect 46 , wherein the method further comprises repeat microstructural models , the diagnostic model meeting a ing ( b ) - ( d ) for all voxels associated with a specified region threshold congruence between the one or more mea of the brain . sured MRI parameters and the one or more simulated [0231 ] 49 . The non - transitory computer- readable medium MRI parameters associated with the diagnostic model ; of aspect 46 , wherein the method further comprises repeat and ing ( b ) - ( d ) for all voxels associated with an entirety of the [ 0218 ] ( d ) for the voxel of the plurality of voxels , using brain . the diagnostic model to determine the disorder state of [0232 ] 50 . The non -transitory computer -readable medium the tissue associated with the voxel . of aspect 46 , wherein the method further comprises repeat [0219 ] 41 . The method of aspect 38 , wherein the tissue is ing ( a ) -( d ) for a plurality of MRI images, each MRI image selected from the group consisting of: spinal cord tissue , of the plurality of MRI images associated with a brain heart tissue, vascular tissue , lung tissue, liver tissue, kidney selected from a plurality of brains, each brain of the plurality tissue , esophageal tissue , stomach tissue, intestinal tissue , of brains associated with a subject selected from a plurality pancreatic tissue , thyroid tissue , adrenal tissue , spleen tissue , of subjects . lymphatic tissue , appendix tissue , breast tissue , bladder [0233 ] 51 . The non - transitory computer- readable medium tissue , vaginal tissue , ovarian tissue, uterine tissue , penile of any one of aspects 42 - 50 , wherein the MRI image is tissue, testicular tissue , prostatic tissue, skeletal muscle selected from the group consisting of: a longitudinal relax tissue , skin , and non -brain tissue of the head and neck . ation time ( T1) -weighted MRI image , a transverse relax [0220 ] 42 . A non -transitory computer- readable medium ation time ( T2 ) - weighted MRI image , and a diffusion comprising machine - executable code that, upon execution weighted MRI image . by one or more computer processors , implements a method [0234 ] 52 . The non - transitory computer- readable medium for detecting a disorder state of brain tissue in a brain of a of any one of aspects 42 - 51 , wherein the measured MRI subject , the method comprising : parameter is selected from the group consisting of: a longi [0221 ] (a ) obtaining magnetic resonance imaging tudinal relaxation time (T1 ) , a transverse relaxation time (MRI ) data comprising at least one MRI image of the ( T2 ) , and a diffusion coefficient. brain , the MRI image comprising a plurality of voxels , [0235 ] 53 . The non -transitory computer- readable medium a voxel of the plurality of voxels being associated with of any one of aspects 42- 52 , wherein the simulated MRI US 2018 /0268942 A1 Sep . 20 , 2018 parameter is selected from the group consisting of: a longi disorder state of the brain tissue associated with the voxel is tudinal relaxation time ( T1 ) , a transverse relaxation time© achieved at an accuracy of at least 90 % . ( T2 ) , and a diffusion coefficient. [ 0250 ] 66 . The non -transitory computer - readable medium [ 02361 54 . The non - transitory computer - readable medium of any one of aspects 48 -65 , wherein determining the of any one of aspects 42 -53 , wherein the one or more disorder state across the brain tissue associated with the microstructural models comprise information regarding a specified region of the brain is achieved at an accuracy of at parameter selected from the group consisting of: intracellu least 90 % . lar content, extracellular content, distribution of extracellu lar content within interstitial space, distribution of intracel [0251 ] 67. The non - transitory computer- readable medium lular content within intracellular space , and tissue geometry. of any one of aspects 49 -66 , wherein determining the [0237 ] 55 . The non -transitory computer -readable medium disorder state of the brain tissue associated with the whole of any one or aspects 42 -54 , wherein the one or more brain of the subject is achieved at an accuracy of at least microstructural models comprise measured or predicted 90 % . values of a parameter selected from the group consisting of: [0252 ] 68 . The non - transitory computer- readable medium cell density , cell shape , cell geometry , cell size , cell distri of any one of aspects 50 -67 , wherein determining the bution , intercellular spacing , extracellular matrix homoge disorder state of the brain tissue associated the plurality of neity , interstitial tortuosity , water to protein ratio , water to subjects is achieved at an accuracy of at least 90 % . lipid ratio , water to carbohydrate ratio , protein to lipid ratio , [0253 ] 69 . The non - transitory computer- readable medium protein to carbohydrate ratio , and lipid to carbohydrate ratio . of any one of aspects 42 -68 , wherein the disorder is a [ 0238 ] 56 . The non - transitory computer - readable medium non - neurodegenerative disorder. of any one of aspects 42 - 55 , wherein the one or more [0254 ] 70 . The non - transitory computer - readable medium microstructural models are selected from a microstructural of aspect 69, wherein the disorder is selected from the group model library . consisting of : a primary neoplasm , a metastatic neoplasm , a [ 0239 ] 57 . The non -transitory computer - readable medium seizure disorder, a seizure disorder with focal cortical dys of aspect 56 , wherein the microstructural model library plasia , a demyelinating disorder , a non -neurodegenerative comprises at least 100 microstructural models . encephalopathy, a cerebrovascular disease , and a psycho 10240 ) 58 . The non -transitory computer -readable medium logical disorder . of aspect 56 or 57 , wherein the microstructuralmodel library [0255 ] 71 . The non - transitory computer- readable medium is constructed by: of any one of aspects 42 -68 , wherein the disorder is a 10241 ] ( a ) creating a first microstructuralmodel corre neurodegenerative disorder . sponding to a brain state that is not associated with a [0256 ] 72 . The non - transitory computer- readable medium disorder , and of aspect 71, wherein the method enables diagnosis of a [0242 ] (b ) iteratively subjecting the first microstructural neurodegenerative disorder more than 5 years prior to the model to a perturbation , each iteration producing an development of symptoms associated with the neurodegen additional perturbed microstructural model. erative disorder . [ 0243] 59 . The non - transitory computer -readable medium [0257 ] 73 . The non - transitory computer- readable medium of aspect 58 , wherein ( b ) comprises subjecting the first of aspect 71 or 72 , wherein the method enables monitoring microstructural model to at least 100 iterations to generate at of the neurodegenerative disorder at a plurality of time least 100 perturbed microstructural models . points, the plurality of time points separated by a plurality of 10244 ] 60 . The non -transitory computer -readable medium time intervals . of aspect 58 or 59, wherein the first microstructural model is selected based on knowledge of the brain region associ [0258 ] 74 . The non - transitory computer -readable medium ated with the voxel. of any one of aspects 71 - 73 , wherein the neurodegenerative [ 0245 ] 61. The non - transitory computer- readable medium disorder is selected from the group consisting of: Alzheim of any one of aspects 58 -60 , wherein the perturbation er' s disease , a non - Alzheimer ' s dementia disorder , Parkin comprises an operation selected from the group consisting son ' s disease , a Parkinsonism disorder, a motor neuron of: depleting cells , altering cellular morphology or distribu disease , Huntington ' s disease , a Huntington ' s disease - like tion , altering intracellular or interstitial physico - chemical syndrome, a transmissible spongiform encephalopathy , composition or distribution , altering extracellular matrix chronic traumatic encephalopathy , and a tauopathy . composition or distribution , and altering intercellular spac [0259 ] 75 . The non - transitory computer- readable medium ing . of any one of aspects 42 -74 , wherein the method further [0246 ] 62 . The non - transitory computer- readable medium comprises constructing a brain map that , for each voxel of of any one of aspects 58 -61 , wherein the perturbation the plurality of voxels , indicates the disorder state of the comprises a stochastic procedure . brain tissue associated with the voxel. [0247 ] 63 . The non -transitory computer -readable medium [ 0260 ] 76 . The non -transitory computer - readable medium of any one of aspects 42 -62 , wherein the threshold congru of aspect 75 , wherein the method further comprises display ence is determined by computing an objective function ing the brain map on a graphical user interface of an between the one or more measured MRI parameters and the electronic device of a user . one or more simulated MRI parameters . [ 0261 ] 77 . The non -transitory computer - readable medium 102481 64 . The non -transitory computer -readable medium of aspect 75 or 76 , wherein the brain map comprises a of aspect 63 , wherein the objective function comprises an L1 qualitative abnormality map . norm or an L2 norm . [0262 ] 78 . The non - transitory computer -readable medium [0249 ] 65 . The non -transitory computer -readable medium of aspect 75 or 76 , wherein the brain map comprises a binary of any one of aspects 42 -64 , wherein determining the abnormality map . US 2018 /0268942 A1 Sep . 20 , 2018

102631 79 . The non - transitory computer -readable medium [0275 ] ii . for the voxel of the plurality of voxels , of aspect 75 or 76 , wherein the brain map comprises a select a diagnostic model from the one or more quantitative abnormality map . microstructural models , the diagnostic model meet [ 02641 80 . The non - transitory computer - readable medium ing a threshold congruence between the one or more of aspect 75 or 76 , wherein the brain map comprises a measured MRI parameters and the one or more percent abnormality map . simulated MRI parameters associated with the diag [0265 ] 81. A non - transitory computer- readable medium nostic model; and comprising machine - executable code that , upon execution [ 0276 ] iii . for the voxel of the plurality of voxels, use by one or more computer processors , implements a method the diagnostic model to determine the disorder state for detecting a disorder state of a tissue of a subject, the of the brain tissue associated with the voxel. method comprising : (0277 ) 84 . The system of aspect 83 , wherein each voxel [ 0266 ] (a ) obtaining magnetic resonance imaging comprises a plurality of measured MRI parameters. (MRI ) data comprising at least one MRI image of the 102781 85 . The system of aspect 83 or 84 , wherein the one tissue , the MRI image comprising a plurality of voxels , or more measured MRI parameters are a plurality of mea a voxel of the plurality of voxels being associated with sured MRI parameters . the tissue of the subject and comprising one or more 02791 86 . The system of any one of aspects 83 - 85 , measured MRI parameters in the MRI data ; wherein the one or more simulated MRI parameters are a [ 0267] (b ) for the voxel of the plurality of voxels , using plurality of simulated MRI parameters . one or more computer processors to process the one or (0280 ] 87 . The system of any one of aspects 83 - 86 , more measured MRI parameters with one or more wherein the one or more computer processors are further simulated MRI parameters for the voxel , the one or individually or collectively programmed to repeat (b ) - ( d ) more simulated MRI parameters being generated from one or more times for additional voxels of the plurality of voxels . one or more microstructural models at the voxel; [0281 ] 88 . The system of aspect 87 , wherein the one or [0268 ] ( c ) for the voxel of the plurality of voxels , more computer processors are further individually or col selecting a diagnostic model from the one or more lectively programmed to repeat ( b ) - ( d ) for all other voxels of microstructural models , the diagnostic model meeting a the plurality of voxels . threshold congruence between the one or more mea [0282 ] 89 . The system of aspect 87 , wherein the one or sured MRI parameters and the one or more simulated more computer processors are further individually or col MRIparameters associated with the diagnostic model ; lectively programmed to repeat (b ) - ( d ) for all voxels asso and ciated with a specified region of the brain . [0269 ] ( d ) for the voxel of the plurality of voxels , using [0283 ] 90 . The system of aspect 87 , wherein the one or the diagnostic model to determine the disorder state of more computer processors are further individually or col the tissue associated with the voxel. lectively programmed to repeat ( b ) - ( d ) for all voxels asso [0270 ] 82 . The non -transitory computer -readable medium ciated with an entirety of the brain . of aspect 82 , wherein the tissue is selected from the group [0284 ] 91. The system of aspect 87 , wherein the one or consisting of: spinal cord tissue , heart tissue, vascular tissue , more computer processors are further individually or col lung tissue , liver tissue , kidney tissue , esophageal tissue , lectively programmed to repeat ( a ) - ( d ) for a plurality of MRI stomach tissue , intestinal tissue, pancreatic tissue , thyroid images, each MRI image of the plurality of MRI images tissue, adrenal tissue , spleen tissue, lymphatic tissue, appen associated with a brain selected from a plurality of brains , dix tissue , breast tissue, bladder tissue , vaginal tissue, ovar each brain of the plurality of brains associated with a subject ian tissue, uterine tissue , penile tissue, testicular tissue , selected from a plurality of subjects . prostatic tissue , skeletal muscle tissue , skin , and non -brain [02851 92 . The system of any one of aspects 83 - 91 , tissue of the head and neck . wherein the MRI image is selected from the group consist [0271 ] 83 . A system for determining a disorder state of ing of: a longitudinal relaxation time ( T1 ) -weighted MRI brain tissue in a brain of a subject, comprising: image , a transverse relaxation time ( T2 ) -weighted MRI [0272 ] ( a ) a database comprising magnetic resonance image , and a diffusion -weighted MRI image . imaging (MRI ) data comprising at least one MRI image [0286 ] 93. The system of any one of aspects 83 - 92 , of the brain , the MRI image comprising a plurality of wherein the measured MRI parameter is selected from the voxels , a voxel of the plurality of voxels being asso group consisting of: a longitudinal relaxation time (T1 ) , a ciated with the brain tissue of the brain of the subject transverse relaxation time ( T2 ), and a diffusion coefficient. and comprising a measured MRI parameter in the MRI [0287 ] 94 . The system of any one of aspects 83 - 93 , data ; and wherein the simulated MRI parameter is selected from the 10273 ] ( b ) one or more computer processors operatively group consisting of: a longitudinal relaxation time ( T1 ) , a coupled to the database , wherein the one or more transverse relaxation time ( T2 ) , and a diffusion coefficient . computer processors are individually or collectively (0288 ] 95 . The system of any one of aspects 83 - 94 , programmed to : wherein the one or more microstructural models comprise [ 0274 ] i . for the voxel of the plurality of voxels , use information regarding a parameter selected from the group one or more computer processors to process the one consisting of : intracellular content, extracellular content, or more measured MRI parameters with one or more distribution of extracellular content within interstitial space , simulated MRI parameters for the voxel, the one or distribution of intracellular content within intracellular more simulated MRI parameters being generated space , and tissue geometry . from one or more microstructural models at the [0289 ] 96 . The system of any one or aspects 83 - 95 , voxel; wherein the one or more microstructural models comprise US 2018 /0268942 A1 Sep . 20 , 2018 measured or predicted values of a parameter selected from neoplasm , a metastatic neoplasm , a seizure disorder, a the group consisting of: cell density , cell shape , cell geom seizure disorder with focal cortical dysplasia , a demyelinat etry , cell size , cell distribution , intercellular spacing, extra ing disorder, a non -neurodegenerative encephalopathy , a cellular matrix homogeneity , interstitial tortuosity, water to cerebrovascular disorder , and a psychological disorder . protein ratio , water to lipid ratio , water to carbohydrate ratio , [0307 ) 112 . The system of any one of aspects 83 - 111 , protein to lipid ratio , protein to carbohydrate ratio , and lipid wherein the disorder is a neurodegenerative disorder . to carbohydrate ratio . [0308 ] 113 . The system of aspect 112 , wherein the system [ 0290 ] 97 . The system of any one of aspects 83 - 96 , enables diagnosis of a neurodegenerative disorder more than wherein the one or more microstructuralmodels are selected 5 years prior to the development of symptoms associated from a microstructural model library. with the neurodegenerative disorder. [0291 ] 98 . The system of aspect 97 , wherein the micro [0309 ] 114 . The system of aspect 112 or 113 , wherein the structural model library comprises at least 100 microstruc system enables monitoring of the neurodegenerative disor tural models . der at a plurality of time points , the plurality of time points [0292 ] 99 . The system of aspect 97 or 98 , wherein the separated by a plurality of time intervals . microstructural model library is constructed by : 10310 ] 115 . The system any one of aspects 112 - 114 . [0293 ] ( a ) creating a first microstructural model corre wherein the neurodegenerative disorder is selected from the sponding to a brain state that is not associated with a group consisting of : Alzheimer ' s disease , a non - Alzheimer ' s disorder ; and dementia disorder , Parkinson ' s disease, a Parkinsonism dis [ 0294 ] (b ) iteratively subjecting the first microstructural order, a motor neuron disease , Huntington ' s disease, a model to a perturbation , each iteration producing an Huntington ' s disease - like syndrome, a transmissible spon additional perturbed microstructural model. giform encephalopathy , chronic traumatic encephalopathy , [0295 ] 100 . The system of aspect 99, wherein ( b ) com and a tauopathy . prises subjecting the first microstructural model to at least [ 0311 ] 116 . The system of any one of aspects 83 -115 , 100 iterations to generate at least 100 perturbed microstruc wherein the one or more computer processors are further tural models . individually or collectively programmed to construct a brain [ 0296 ] 101. The system of aspect 99 or 100 , wherein the map that, for each voxel of the plurality of voxels , indicates first microstructural model is selected based on knowledge the disorder state of the brain tissue associated with the of the brain region associated with the voxel. voxel. [ 0297] 102 . The system of any one of aspects 99 - 101 , [0312 ] 117 . The system of aspect 116 , wherein the one or wherein the perturbation comprises an operation selected more computer processors are further individually or col from the group consisting of: depleting cells , altering cel lectively programmed to display the brain map on a graphi lular morphology or distribution , altering intracellular or cal user interface of an electronic device of a user. interstitial physico - chemical composition or distribution , [0313 ] 118 . The system of aspect 116 or 117, wherein the altering extracellular matrix composition or distribution , and brain map comprises a qualitative abnormality map . altering intercellular spacing . [0314 ] 119 . The system of aspect 116 or 117 , wherein the [0298 ] 103. The system of any one of aspects 99- 102 , brain map comprises a binary abnormality map . wherein the perturbation comprises a stochastic procedure . [0315 ] 120 . The system of aspect 116 or 117 , wherein the [0299 ] 104 . The system of any one of aspects 83 - 103, brain map comprises a quantitative abnormality map . wherein the threshold congruence is determined by comput [0316 ] 121. The system of aspect 116 or 117 , wherein the ing an objective function between the one or more measured brain map comprises a percent abnormality map . MRI parameters and the one or more simulated MRI param [0317 ] 122 . A system for determining a disorder state of a eters . tissue in a portion of a body of a subject, comprising : [ 0300 ] 105 . The system of aspect 104 , wherein the objec [ 0318 ] (a ) a database comprising magnetic resonance tive function comprises an Li norm or an L2 norm . imaging (MRI ) data comprising at least one MRI image [ 0301] 106 . The system of any one of aspects 83 - 105 , of the brain , the MRI image comprising a plurality of wherein determining the disorder state of the brain tissue voxels , a voxel of the plurality of voxels being asso associated with the voxel is achieved at an accuracy of at ciated with the brain tissue of the brain of the subject least 90 % . and comprising a measured MRI parameter in the MRI [ 0302 ] 107 . The system of any one of aspects 89 - 106 , data ; and wherein determining the disorder state across the brain [0319 ] ( b ) one or more computer processors operatively tissue associated with the specified region of the brain is coupled to the database , wherein the one or more achieved at an accuracy of at least 90 % . computer processors are individually or collectively [ 0303] 108. The system of any one of aspects 90 - 107, programmed to : wherein determining the disorder state of the brain tissue [0320 ] i. for the voxel of the plurality of voxels, use associated with the whole brain of the subject is achieved at one or more computer processors to process the one an accuracy of at least 90 % . or more measured MRIparameters with one or more [0304 ] 109 . The system of any one of aspects 91 - 108 , simulated MRI parameters for the voxel, the one or wherein determining the disorder state of the brain tissue more simulated MRI parameters being generated associated the plurality of subjects is achieved at an accu from one or more microstructural models at the racy of at least 90 % . voxel ; [0305 ] 110 . The system of any one of aspects 83 - 109 , [0321 ] ii . for the voxel of the plurality of voxels , wherein the disorder is a non -neurodegenerative disorder. select a diagnostic model from the one or more [ 0306 ] 111. The system of aspect 110 , wherein the disor microstructural models , the diagnostic model meet der is selected from the group consisting of: a primary ing a threshold congruence between the one or more US 2018 /0268942 A1 Sep . 20 , 2018 24

measured MRI parameters and the one or more 10 . The method of claim 1 , wherein the MRI image is simulated MRI parameters associated with the diag selected from the group consisting of: a longitudinal relax nostic model ; and ation time ( T1 ) -weighted MRI image , a transverse relax [ 0322 ] iii. for the voxel of the plurality of voxels , use ation time ( T2 ) -weighted MRI image, and a diffusion the diagnostic model to determine the disorder state weighted MRI image . of the tissue associated with the voxel. 11 . The method of claim 10 , wherein the measured MRI [0323 ] 123 . The system of aspect 122 , wherein the tissue parameter is selected from the group consisting of: a longi is selected from the group consisting of : spinal cord tissue , tudinal relaxation time ( T1 ) , a transverse relaxation time heart tissue , vascular tissue, lung tissue , liver tissue , kidney ( T2 ) , and a diffusion coefficient. tissue , esophageal tissue , stomach tissue , intestinal tissue , 12 . ( canceled ) pancreatic tissue , thyroid tissue , adrenal tissue , spleen tissue , 13 . The method of claim 1 , wherein the one or more lymphatic tissue , appendix tissue , breast tissue , bladder microstructural models comprise information regarding a tissue, vaginal tissue , ovarian tissue , uterine tissue , penile parameter selected from the group consisting of: intracellu tissue , testicular tissue , prostatic tissue , skeletal muscle lar content, extracellular content , distribution of extracellu tissue, skin , and non -brain tissue of the head and neckek . lar content within interstitial space , distribution of intracel 1 . A method for determining a disorder state of brain lular content within intracellular space , and tissue geometry . tissue in a brain of a subject, comprising : 14 . The method of claim 13 , wherein the one or more ( a ) obtaining magnetic resonance imaging (MRI ) data microstructural models comprise measured or predicted comprising at least one MRI image of the brain , the values of a parameter selected from the group consisting of: MRI image comprising a plurality of voxels , a voxel of cell density , cell shape , cell geometry, cell size , cell distri the plurality of voxels being associated with the brain bution , intercellular spacing , extracellular matrix homoge tissue of the brain of the subject and comprising one or neitv , interstitial tortuosity , water to protein ratio , water to more measured MRI parameters in the MRI data ; lipid ratio , water to carbohydrate ratio , protein to lipid ratio , ( b ) for the voxel of the plurality of voxels , using one or protein to carbohydrate ratio , and lipid to carbohydrate ratio . more computer processors to process the one ormore 15 . The method of claim 14 , wherein the one or more measured MRI parameters with one or more simulated microstructural models are selected from a microstructural MRI parameters for the voxel, the one or more simu model library . lated MRI parameters being generated from one or 16 . ( canceled ) more microstructural models at the voxel; 17 . The method of claim 15 , wherein the microstructural ( c ) for the voxel of the plurality of voxels, selecting a model library is constructed by : diagnostic model from the one or more microstructural models , the diagnostic modelmeeting a threshold con ( a ) creating a first microstructuralmodel corresponding to gruence between the one or more measured MRI a brain state that is not associated with a disorder ; and parameters and the one or more simulated MRI param (b ) iteratively subjecting the first microstructural model to eters associated with the diagnostic model ; and a perturbation , each iteration producing an additional ( d ) using the diagnostic model to determine the disorder perturbed microstructural model. state of the brain tissue associated with at least the 18 . (canceled ) voxel . 19 . (canceled ) 2 . ( canceled ) 20 . The method of claim 17, wherein the perturbation 3 . The method of claim 1 , wherein the one or more comprises an operation selected from the group consisting measured MRI parameters are a plurality of measured MRI of: depleting cells , altering cellular morphology or distribu parameters tion , altering intracellular or interstitial physico -chemical 4 . The method of claim 3 , wherein the one or more composition or distribution , altering extracellular matrix simulated MRI parameters are a plurality of simulated MRI composition or distribution , and altering intercellular spac parameters. ing . 5 . ( canceled ) 21 . The method of claim 20 , wherein the perturbation 6 . The method of claim 1 , further comprising repeating comprises a stochastic procedure . ( b ) - ( d ) for all other voxels of the plurality of voxels . 22 . The method of claim 21 , wherein the threshold 7 . The method of claim 1 , further comprising repeating congruence is determined by computing an objective func ( b ) - ( d ) for all voxels associated with a specified region of the tion between the one or more measured MRI parameters and brain to determine disorder states across the brain tissue the one or more simulated MRI parameters . associated with the specified region of the brain of the 23 . The method of claim 22 , wherein the objective func subject . tion comprises an L1 norm or an L2 norm . 8. The method of claim 1 , further comprising repeating 24 . The method of claim 1, wherein determining the ( b ) - ( d ) for all voxels associated with an entirety of the brain disorder state of the brain tissue associated with the voxel is to determine disorder states of the brain tissue associated achieved at an accuracy of at least 90 % . with the entirety of the brain of the subject. 25 . The method of claim 7 , wherein determining the 9 . The method of claim 1, further comprising repeating disorder states across the brain tissue associated with the (a ) - ( d ) for a plurality of MRI images , each MRI image of thede specified region of the brain is achieved at an accuracy of at plurality of MRI images associated with a brain selected least 90 % . from a plurality of brains , each brain of the plurality of 26 . The method of claim 8 , wherein determining the brains associated with a subject selected from a plurality of disorder states of the brain tissue associated with the entirety subjects , to determine disorder states of the brain tissue of the brain of the subject is achieved at an accuracy of at associated the plurality of subjects . least 90 % . US 2018 /0268942 A1 Sep . 20 , 2018 25

27 . The method of claim 9 , wherein determining the Alzheimer 's disease , a non - Alzheimer' s dementia disorder, disorder states of the brain tissue associated with the plu - Parkinson ' s disease , a Parkinsonism disorder , a motor neu rality of subjects is achieved at an accuracy of at least 90 % . ron disease, Huntington 's disease, a Huntington ' s disease 28 . The method of claim 1 , wherein the disorder is a like syndrome, transmissible spongiform encephalopathy, non -neurodegenerative disorder . chronic traumatic encephalopathy, and a tauopathy . 29 . The method of claim 28 , wherein the disorder is 34 . The method of claim 1, further comprising construct selected from the group consisting of: a primary neoplasm , ing a brain map that, for each voxel of the plurality of voxels , a metastatic neoplasm , a motor neuron disease , a seizure indicates the disorder state of the brain tissue associated with disorder, a seizure disorder with focal cortical dysplasia , the voxel. multiple sclerosis , a non -neurodegenerative encephalopathy , 35 . The method of claim 34 , further comprising display and a psychological disorder. ing the brain map on a graphical user interface of an 30 . The method of claim 1 , wherein the disorder is a electronic device of a user. neurodegenerative disorder. 36 . The method of claim 34 , wherein the brain map is 31 . The method of claim 30 , wherein the method enables selected from the group consisting of: a qualitative abnor diagnosis of a neurodegenerative disorder more than 5 years mality map , a binary abnormality map , a quantitative abnor prior to the development of symptoms associated with the mality map , and a percent abnormality map . neurodegenerative disorder. 32 . The method of claim 30 , wherein the method enables 37 . (canceled ) monitoring of the neurodegenerative disorder at a plurality 38 . (canceled ) of time points , the plurality of time points separated by a 39 . (canceled ) plurality of time intervals . 40 . ( canceled ) 33 . The method of claim 30 , wherein the neurodegenera 41 . (canceled ) tive disorder is selected from the group consisting of: * * * * *