Non-Invasive Fetal Electrocardiogram Monitoring Techniques: Potential and Future Research Opportunities in Smart Textiles

signals

Review Non-Invasive Fetal Electrocardiogram Monitoring Techniques: Potential and Future Research Opportunities in Smart Textiles

Geetika Aggarwal * and Yang Wei

Smart Wearable Research Group, Department of Engineering, Nottingham Trent University, Nottingham NG11 8NS, UK; [email protected] * Correspondence: [email protected]

Abstract: During the pregnancy, fetal electrocardiogram (FECG) is deployed to analyze fetal heart rate (FHR) of the fetus to indicate the growth and health of the fetus to determine any abnormalities and prevent diseases. The fetal electrocardiogram monitoring can be carried out either invasively by placing the electrodes on the scalp of the fetus, involving the skin penetration and the risk of infection, or non-invasively by recording the fetal heart rate signal from the mother’s abdomen through a placement of electrodes deploying portable, wearable devices. Non-invasive fetal electrocardiogram (NIFECG) is an evolving technology in fetal surveillance because of the comfort to the pregnant women and being achieved remotely, speciﬁcally in the unprecedented circumstances such as pandemic or COVID-19. Textiles have been at the heart of human technological progress for thousands of years, with textile developments closely tied to key inventions that have shaped societies. The relatively recent invention of smart textiles is set to push boundaries again and has already opened the potential for garments relevant to medicine, and health monitoring. This paper aims to discuss the different technologies and methods used in non-invasive fetal electrocardiogram Citation: Aggarwal, G.; Wei, Y. (NIFECG) monitoring as well as the potential and future research directions of NIFECG in the smart Non-Invasive Fetal textiles area. Electrocardiogram Monitoring Techniques: Potential and Future Keywords: non-invasive FECG; fetal monitoring; healthcare Research Opportunities in Smart Textiles. Signals 2021, 2, 392–412. https://doi.org/10.3390/signals2030025

1. Introduction Academic Editor: Yoshiharu Yamamoto Electrocardiogram (ECG) can be defined as a graphical representation of bioelectrical signals helpful in determining the functionality of the heart through the analysis of graphic Received: 21 February 2021 representation obtained during the measurement of cardiac cycle of the person or human Accepted: 23 June 2021 body. The aim of the biomedical research is to continuously improve the diagnostic devices Published: 29 June 2021 and develop non-invasive methods of health monitoring, in addition to upgrading already existing devices, thus reducing the cost involved [1–6]. The ongoing research predicts that Publisher’s Note: MDPI stays neutral the main cause of prenatal death is heart defects, as congenital heart defects easily occur with regard to jurisdictional claims in during the formation of the heart at the initial stage of pregnancy [7–10]. published maps and institutional affil- During the pregnancy, the monitoring of fetal heart rate is essential to identify the iations. proper supply of oxygen, nutrients, and growth of the fetus. Monitoring of the fetus during pregnancy may help in recognizing the pathological conditions, such as fetal hypoxia, allowing prompt medical interventions before irreversible changes taking place. The abnormality in the fetal heart rate of the fetus indicates that there is insufficient oxygen Copyright: © 2021 by the authors. supplied or other problems to the fetus. Licensee MDPI, Basel, Switzerland. FHR during pregnancy or labor can be monitored through invasive and non-invasive This article is an open access article fetal methods [11,12] so that the heart functionality of the fetus can be predicted to reflect distributed under the terms and the growth and wellbeing of the fetus [13–15]. Electrocardiography (ECG) was used conditions of the Creative Commons by researchers in 1906 to invasively determine the FHR through abdominal during the Attribution (CC BY) license (https:// pregnancy [15,16]. In 1958, Edward H. Hon used the successive R waves method from ECG creativecommons.org/licenses/by/ for the calculation and estimation of FHR non-invasively [17–22]. With the advancement in 4.0/).

Signals 2021, 2, 392–412. https://doi.org/10.3390/signals2030025 https://www.mdpi.com/journal/signals Signals 2021, 2 FOR PEER REVIEW 2

researchers in 1906 to invasively determine the FHR through abdominal during the preg- Signals 2021, 2 393 nancy [15,16]. In 1958, Edward H. Hon used the successive R waves method from ECG for the calculation and estimation of FHR non-invasively [17–22]. With the advancement in technology, Callagan in 1964 monitored the FHR non-invasively using the doppler technology,sound monitoring Callagan procedure, in 1964 monitored in which an the ultrasound FHR non-invasively beam was usingsent through the doppler the abdo- sound monitoringmen of the procedure,mother and in the which reflected an ultrasound signal was beam measured, was sent thus through helping the to abdomen analyze ofFHR the mother[23–27]. and the reflected signal was measured, thus helping to analyze FHR [23–27]. UnlikeUnlike invasiveinvasive fetal electrocardiography (FECG), (FECG), which which involves involves surgery surgery to to insert insert thethe electrodeselectrodes intointo the scalp of the fetus through through the the abdomen abdomen of of pregnant pregnant women, women, non- non- invasiveinvasive fetalfetal electrocardiographyelectrocardiography (NIFECG) (NIFECG) deploys deploys electrodes electrodes placed placed on on the the abdomen abdomen of theof the pregnant pregnant mother mother to to extract extract the the fetus’ fetus’ health health information. information. The electrical signal signal of of the the fetalfetal heartheart raterate helpshelps identifyidentify the development of of the the fetus fetus in in addition addition to to the the presence presence of of anyany congenitalcongenital heartheart disease within the fetus [28–31]. [28–31]. ThisThis paper is is divided divided into into five five sections. sections. Section Section 2 provides2 provides an overview an overview of fetal of elec- fetal electrocardiographytrocardiography (FECG). (FECG). The Thedifferent different techniques techniques to obtain to obtain FECG FECG non-invasively non-invasively are pre- are presentedsented in Section in Section 3, where3, where they they are are critically critically discussed discussed and and compared. compared. The The importance importance of ofFECG FECG and and the the electrode electrode configurations configurations for for FECG FECG signal signal extraction extraction are are listed listed in in Section Section 4.4 . SectionSection5 5 lists lists the the conclusions, conclusions, andand presentspresents thethe potentialpotential futurefuture research directions.

2.2. BackgroundBackground of FECG FetalFetal electrocardiogramelectrocardiogram (FECG) is is a biom biomedicaledical signal, signal, see see the the example example shown shown in in FigureFigure1 ,1, that that provides provides an an electricalelectrical representationrepresentation ofof fetalfetal heartheart rate (FHR). It It is is composed composed ofof threethree parts:parts: (i)(i) the the P-wave, P-wave, (ii) (ii) the the QRS-complex, QRS-complex, which which is is associated associated with with the the contraction contrac- oftion the of ventricles, the ventricles, and and due due to theto the magnitude magnitud ofe of the the R-wave R-wave itit isis extremely reliable, reliable, and and (iii)(iii) thethe T-wave,T-wave, whichwhich correspondscorresponds to the repolarization phase phase and and follows follows each each heart heart contractioncontraction [31–36]. [31–36]. The The delay delay associated associated to tothe the R-R R-R interval interval leads leads to the to heartbeat’s the heartbeat’s fre- frequencyquency and and provides provides useful useful information information about about the theheart heart condition. condition. Morphologies Morphologies of in- of interestterest include include the the shape, shape, size, size, and and duration duration of of individual individual and and groups groups of of FECG FECG waveforms waveforms asas wellwell asas thethe variousvarious ratios relating these quantities to to each each other other [19,37–45]. [19,37–45].

FigureFigure 1.1.Key Key features features of of FECG FECG [45 [45]] (Reprinted (Reprinted with with permission permission from from ref [45 ref] Radek [45] Radek Martinek Martinek (2012)). (2012)). The FECG extracted from the maternal electrocardiogram (MECG) is exceedingly small,The about FECG 5 times extracted less in from amplitude the maternal compared electrocardiogram to the MECG and (MECG) is sometimes is exceedingly embedded insmall, the noiseabout signals 5 times [19 less,42 in–46 amplitude]. compared to the MECG and is sometimes embedded Asin the shown noise in signals Figure 2[19,42–46]., the FECG signals are usually buried in noise and artefacts, such as interferenceAs shown noise,in Figure muscle 2, the contractions, FECG signals instrumental are usually noise, buried etc., in whichnoise and easily artefacts, corrupt thesuch FECG as interference signal [47– noise,49]. Therefore, muscle contractions, signal processing instrumental is essential noise, to etc., remove which the easily noise cor- and artefactsrupt the FECG to extract signal the [47–49]. actual Therefore, FECG signal signal to analyzeprocessing and is determineessential to theremove growth the noise of the fetus [50].

Signals 2021, 2 FOR PEER REVIEW 3

and artefacts to extract the actual FECG signal to analyze and determine the growth of the fetus [50]. Signals 2021, 2 394 and artefacts to extract the actual FECG signal to analyze and determine the growth of the fetus [50].

Figure 2. The example of FECG embedded within the MECG signal [45] (Reprinted with permission from ref [45] Radek Martinek (2012)). FigureFigure 2. 2. TheThe example example of of FECG FECG embedded embedded within within the the MECG MECG signal signal [45] [45] (Reprinted (Reprinted with with permission permission 3. Comparison of Fetal Heart Rate Monitoring Techniques fromfrom ref ref [45] [45] Radek Radek Martinek Martinek (2012)). (2012)). The fetal heart rate or fetal electrocardiography monitoring thorough abdominal FECG3.3. Comparison Comparison electrodes, of of i.e., Fetal Fetal through Heart Heart Ratenon-invasive Rate Monitoring Monitoring fetal Techniques Techniqueselectrocardiogra m (NIFECG), has wide- spreadTheThe clinical fetal heartheartacceptance rate rate or oras fetal fetalit is electrocardiography suitableelectrocar fordiography long-term monitoring monitoringrecordings thorough withoutthorough abdominal any abdominal surgery, FECG unlikeFECGelectrodes, electrodes,invasive i.e., through FECG. i.e., through This non-invasive section non-invasive illustrates fetal electrocardiogram fetal the electrocardiogra commonly (NIFECG), usedm fetal (NIFECG), has heart widespread rate has moni-wide- clin- toringspreadical acceptance techniques. clinical acceptance as it is suitable as it for is long-termsuitable for recordings long-term without recordings any surgery, without unlike any surgery, invasive unlikeFECG.This invasive This section section FECG. illustrates illustrates This the section the commonly commonly illustrates used used the non-invasive fetalcommonly heart rate usedfetal monitoring heartfetal heartrate techniques.monitoring rate moni- techniquestoringThis techniques. section in addition illustrates to FECG. the commonly used non-invasive fetal heart rate monitoring techniquesThis section in addition illustrates to FECG. the commonly used non-invasive fetal heart rate monitoring 3.1.techniques Photoplethysmography in addition to (PPG)FECG. 3.1. Photoplethysmography (PPG) Photoplethysmography (PPG) is used to measure the fetal heart rate, introduced in 1930.3.1. Photoplethysmography PhotoplethysmographyPPG is a non-invasive (PPG) fetal (PPG) monitoring is used toprocedure measure which the fetal uses heart changes rate, in introduced blood level in 1930. PPG is a non-invasive fetal monitoring procedure which uses changes in blood volumesPhotoplethysmography to measure FHR [51–54]. (PPG) The is usedprocedure to measure involves the the fetal detection heart rate, of FHR introduced by trans- in level volumes to measure FHR [51–54]. The procedure involves the detection of FHR by mitting1930. PPG a wavelength is a non-invasive of 650–950 fetal nmmonitoring through procedurethe maternal which abdomen, uses changes having ina peak blood wave- level transmitting a wavelength of 650–950 nm through the maternal abdomen, having a peak length of 890 nm to penetrate the human tissue. The reflected light is analyzed to calculate volumeswavelength to measure of 890 nm FHR to [51–54]. penetrate The the procedure human tissue. involves The the reﬂected detection light of FHR is analyzed by trans- to the fetal heart rate. The example of a PPG signal comprising of both AC and DC compo- mittingcalculate a wavelength the fetal heart of rate.650–950 The nm example through of athe PPG maternal signal ab comprisingdomen, having of both a peak AC and wave- DC nents is shown in Figure 3 [53]. lengthcomponents of 890 isnm shown to penetrate in Figure the3 [human53]. tissue. The reflected light is analyzed to calculate the fetal heart rate. The example of a PPG signal comprising of both AC and DC components is shown in Figure 3 [53].

FigureFigure 3. 3. PPGPPG signal signal comprising comprising of of AC AC and and DC DC components components [53] [53] (Reprinted (Reprinted with with permission permission from from refref (53) [53] Javier Javier Pérez Pérez Contonente). Contonente). FigureThe 3. PPG AC signal component comprising is the of AC resultant and DC of components pulsatile changes[53] (Reprinted in arterial with permission blood volume, from refwhereas (53) Javier the Pérez DC componentContonente). corresponds to the tissues and muscles [53–55]. The AC component is helpful in determining the FHR as the arterial blood volume is synchronous with the heartbeat, but the signal obtained from the AC component is exceedingly small in range due to noises and artefacts [56,57]. The major challenge in determining fetal heart rate through PPG is the extraction of signals from noisy transducer data, which is mainly

Signals 2021, 2 FOR PEER REVIEW 4

The AC component is the resultant of pulsatile changes in arterial blood volume, whereas the DC component corresponds to the tissues and muscles [53–55]. The AC component is helpful in determining the FHR as the arterial blood volume is synchronous Signals 2021, 2 395 with the heartbeat, but the signal obtained from the AC component is exceedingly small in range due to noises and artefacts [56,57]. The major challenge in determining fetal heart rate through PPG is the extraction of signals from noisy transducer data, which is mainly affectedaffected by by acoustics, acoustics, fetal fetal and and maternal maternal contractions, contractions, fetal fetal movements, movements, and and maternal maternal breathing.breathing. The The authors authors of of [52] [52 ]introduced introduced a aPPG PPG system system for for fetal fetal heart heart rate rate monitoring monitoring involvinginvolving a abandpass bandpass filter ﬁlter (BFP) (BFP) in in the the range range of of 710–740 710–740 Hz Hz from from the the abdomen abdomen of of the the mother,mother, a acut-off cut-off frequency frequency of of the the reference reference sign signalal kept kept at at 15 15 Hz, Hz, and and the the signal signal is isdown- down- sampledsampled to to 55 55 Hz, Hz, followed followed by bya bandpass a bandpass filter ﬁlter of 0.6–15 of 0.6–15 Hz to Hz remove to remove artifacts artifacts and un- and wantedunwanted noise noise in the in received the received signal, signal, deployin deployingg four phonogram four phonogram sensors, sensors, as shown as shown in Fig- in ureFigure 4a, and4a, andthe PPG the PPG sensors sensors for experiment for experiment validation, validation, shown shown in Figure in Figure 4b [52].4b [52 ].

(a) (b)

FigureFigure 4. 4.(a()a A) Aschematic schematic diagram diagram showing showing the the locations locations of of phonogram phonogram sensors. sensors. Ch1, Ch1, Ch2, Ch2, Ch3, Ch3, and and Ch4Ch4 are are four four phonogram phonogram sensors.sensors. ((bb)) LocationsLocations of of phonocardiogram phonocardiogram sensors sensors and and fetal fetal ECG ECG electrodes elec- trodesfor the for cross-validation the cross-validation experiment experiment [52] (Reprinted [52] (Reprinted with permission with permission from ref. from [52 ref.] (2018, [52] CC((2018, BY 4.0)).CC BY 4.0)). The FECG signals were captured invasively using PPG electrodes, and the experiment procedureThe FECG and signals signal processingwere captured successfully invasively removed using PPG the noiseelectrodes, and maternal and the artefactsexperi- mentfrom procedure the FECG and signal signal [52, 58processing]. successfully removed the noise and maternal artefacts from the FECG signal [52,58]. 3.2. Cardiotocography (CTG) 3.2. CardiotocographyCardiotocography (CTG) (CTG) is a procedure for fetal heart rate detection introduced by KonradCardiotocography Hammacher in(CTG) 1970 is [59 a– procedure62]. CTG canfor fetal be carried heart rate out bothdetection invasively introduced and non-by Konradinvasively, Hammacher as shown in in 1970 Figure [59–62].5a [ 49 CTG] and can Figure be 5carriedb [ 49]. out In invasiveboth invasively CTG, an and electrode non- invasively,is directly as attached shown toin theFigure scalp 5a of[49] the and fetus Figure using 5b a[49]. wire In through invasive the CTG, cervix. an electrode In contrast, is directlyin non-invasive attached to CTG, the electrodesscalp of the are fetus placed using on a the wire abdomen through of the pregnant cervix. In women contrast, for thein non-invasivefetal heart rate CTG, detection. electrodes The are signal placed quality on the of invasiveabdomen CTG of pregnant is more accuratewomen for but the with fetal less heartcomfort rate anddetection. higher riskThe comparedsignal quality to non-invasive of invasive CTGCTG [is49 ,more62]. accurate but with less comfortSeveral and higher studies risk concluded compared that to long-termnon-invasive continuous CTG [49,62]. monitoring using CTG resulted in discomfort to the patients. In addition, a study comprised of pregnant women at 20 weeks of pregnancy showed that the continuous fetal monitoring using CTG could result in an increase in the number of caesarean sections and misidentiﬁcation of maternal heart rate [61,62].

SignalsSignals 20212021, 2 FOR, 2 PEER REVIEW 5396

(a)

(b)

FigureFigure 5. ( 5.a) (Non-invasivea) Non-invasive FECG FECG measurement. measurement. (b) (bInvasive) Invasive FECG FECG measurement measurement [45] [45 (Reprinted] (Reprinted withwith permission permission from from Radek Radek Martinek Martinek (2012). (2012)).

3.3.Several Doppler studies Sound concluded that long-term continuous monitoring using CTG resulted in discomfortThe Doppler to the patients. principle In is addition, used in fetal a study heart comprised rate monitoring of pregnant by analyzing women theat 20 me- weekschanical of pregnancy activity of showed the heart, that such the continuo as movementus fetal of monitoring valves of the using heart CTG of thecould fetus result in the in ancardiac increase cycle in [the63, 64number]. The of use caesarean of Doppler sections ultrasound and misidentification to monitor the of FHR, maternal as shown heart in rateFigure [61,62].6[64 ] , was proposed in the 1960s, but it is not suitable for continuous monitoring due to discomfort and irritation caused to the patients due to the gel and conduction paste 3.3.used Doppler during Sound the procedure. TheThe Doppler procedure principle is non-invasive is used in fetal and hear deployst rate themonitoring ultrasound by analyzing transducer the placed mechan- on the icalmaternal activity of abdomen. the heart, The such ultrasound as movement with of the valves frequency of the range heart ofof 1–2.3the fetus MHz inpenetrates the cardiac the cycletissue [63,64]. and theThe reﬂected use of Doppl sounder wave ultrasound generated to bymonitor the internal the FHR, structure as shown of the in body Figure or fetus 6 [64],is receivedwas proposed by the in ultrasound the 1960s, transducer, but it is not which suitable analyzes for continuous the fetal heart monitoring rate [65, 66due]. to discomfort and irritation caused to the patients due to the gel and conduction paste used during the procedure.

Signals 2021, 2 FOR PEER REVIEW 6

Signals 2021, 2 397 Signals 2021, 2 FOR PEER REVIEW 6

Figure 6. An illustration of the Doppler sound procedure for FHR monitoring.

The procedure is non-invasive and deploys the ultrasound transducer placed on the maternal abdomen. The ultrasound with the frequency range of 1–2.3 MHz penetrates the tissue and the reflected sound wave generated by the internal structure of the body or FigurefetusFigure is 6.6. received AnAn illustrationillustration by the ofof ultrasound thethe DopplerDoppler transducer, soundsound procedureprocedure which forfor analyzes FHRFHR monitoring. monitoring. the fetal heart rate [65,66].

3.4.3.4. FetalelectrocardiographyThe procedure is non-invasive (FECG)(FECG) and deploys the ultrasound transducer placed on the maternal abdomen. The ultrasound with the frequency range of 1–2.3 MHz penetrates the ForFor the fetus, fetus, fetal fetal electrocardiography electrocardiography (FECG) (FECG) helps helps to analyze to analyze FHR FHR and andhealth health con- tissue and the reflected sound wave generated by the internal structure of the body or conditionsditions because because the heart the heart of the of fetus the is fetus the fir isst the organ first to organ develop to during develop the during first 3–4 the weeks first fetus is received by the ultrasound transducer, which analyzes the fetal heart rate [65,66]. 3–4of pregnancy. weeks of pregnancy. To determine To determine FHR of the FHR fetus, of the several fetus, several fetal monitoring fetal monitoring techniques, techniques, both bothinvasive invasive and andnon-invasive, non-invasive, are areused. used. However, However, considering considering the the comfort comfort of ofthe the patient, patient, a 3.4. Fetalelectrocardiography (FECG) anon-invasive non-invasive process process for for FECG FECG is is usually usually prefe preferredrred [[67–71].67–71]. In [[72],72], Clifford etet al.al. analyzedanalyzed andand evaluatedevaluatedFor the fetus, thethe FECGFECGfetal electrocardiography device’sdevice’s accuracyaccuracy inin (FECG) thethe contextcontext helps ofof to fetalfetal analyze scalpscalp FHR electrodeselectrodes and health (FSE)(FSE) con- byby comparingcomparingditions because thethe resultsresults the heart fromfrom of the the fetus experiment.experiment. is the fir Thirty-twostThirty-two organ to develop pregnantpregnant during womenwomen the were werefirst 3–4 involvedinvolved weeks for data recording and the electrode placement was achieved following the configuration forof pregnancy.data recording To determineand the electrode FHR of placem the fetus,ent wasseveral achieved fetal monitoring following the techniques, configuration both shown in Figure7. The signal quality varied according to fetal positioning. However, the showninvasive in andFigure non-invasive, 7. The signal are quality used. variedHowever, according considering to fetal the positioning. comfort of However,the patient, the a results of the study depicted that an average correlation of 0.96 was achieved in comparison resultsnon-invasive of the studyprocess depicted for FECG that is usuallyan average prefe correlationrred [67–71]. of 0.96 In [72], was Clifford achieved et al.in compari-analyzed to results obtained using invasive fetal scalp electrodes (FSE). sonand toevaluated results obtained the FECG using device’s invasive accuracy fetal inscalp the electrodescontext of (FSE).fetal scalp electrodes (FSE) by comparing the results from the experiment. Thirty-two pregnant women were involved for data recording and the electrode placement was achieved following the configuration shown in Figure 7. The signal quality varied according to fetal positioning. However, the results of the study depicted that an average correlation of 0.96 was achieved in comparison to results obtained using invasive fetal scalp electrodes (FSE).

FigureFigure 7.7. ElectrodeElectrode placementplacement forfor FECG.FECG.

In [69], the authors used a commercially available FECG device AN24 to continuously monitor the fetus and to determine the fetus health. Some of the commercially available

NIFECG devices are shown in Figure8. Figure 7. Electrode placement for FECG.

Signals 2021, 2 FOR PEER REVIEW 7

Signals 2021, 2 In [69], the authors used a commercially available FECG device AN24 to continu-398 ously monitor the fetus and to determine the fetus health. Some of the commercially available NIFECG devices are shown in Figure 8.

(a) (b)

(e)

FigureFigure 8. Examples 8. Examples of ofCE/FDA-certified CE/FDA-certiﬁed commercially commercially available NIFECG-basedNIFECG-based devices devices [69 [69].]. (a )( Mon-a) Monicaica AN24, AN24, (b (b)) Monica Monica Novii Novii Wireless Wireless Patch Patch System, System, ( c(c)) MERIDIAN MERIDIAN M110 M110 FetalFetal MonitoringMonitoring System,Sys- tem, (d) PUREtrace, (e) Nemo Fetal Monitoring System (Reprinted with permission from Radek (d) PUREtrace, (e) Nemo Fetal Monitoring System (Reprinted with permission from Radek Martinek Martinek (2019) (2019, Creative Commons License). (2019) (2019, Creative Commons License)).

To understand the importance of electrode placement, a study was non-invasively conducted using the Monica AN24 Monitor with ﬁve electrodes [41,69,72–74]. It was observed that the non-invasive FECG was a preferred method to provide comfort to pregnant women in comparison to invasive FECG [72,75,76]. However, the results indicated that the signal quality of the FECG data obtained was affected by maternal cardiograph Signals 2021, 2 FOR PEER REVIEW 8

To understand the importance of electrode placement, a study was non-invasively conducted using the Monica AN24 Monitor with five electrodes [41,69,72–74]. It was ob- Signals 2021 2 , served that the non-invasive FECG was a preferred method to provide comfort to preg-399 nant women in comparison to invasive FECG [72,75,76]. However, the results indicated that the signal quality of the FECG data obtained was affected by maternal cardiograph (MECG)(MECG) artefacts artefacts and and noise, noise, bu butt the the FECG FECG signal signal can can bebe extracted extracted after after applying applying a set a setof filtersof ﬁlters [42,77–79]. [42,77–79 In]. [7], In a [7 study], a study was wasconducte conductedd using using three threepairs pairsof electrodes of electrodes which which were placedwere placed on the on abdomen the abdomen of pregnant of pregnant wome women.n. Complex Complex continuous continuous wavelet wavelet transform transform (CCWT)(CCWT) was was deployed deployed for for the the extraction extraction of of th thee FECG FECG signal, signal, using using a afour-stage four-stage procedure, procedure, asas shown shown in in Figure Figure 99..

FigureFigure 9. 9. TheThe proposed proposed 4-stage 4-stage fetal fetal heart heart ra ratete (FHR) (FHR) extraction extraction method method [7]. [7].

InIn the the first first stage, stage, the the signals signals coming coming from from the the three three pairs pairs of of electrodes electrodes were were averaged, averaged, followedfollowed by by stage stage two two comprising comprising of ofCCWT CCWT and and MECG MECG detection. detection. The The detection detection and andex- tractionextraction of FECG of FECG was was carried carried out outin stage in stage three, three, while while stage stage four four focused focused on analyzing on analyzing the the fetal heart rate (FHR). Furthermore, a method to extract FECG from the MECG using a fetal heart rate (FHR). Furthermore, a method to extract FECG from the MECG using a multivariate singular spectrum analysis (MSSA) was used, comprising of two stages known multivariate singular spectrum analysis (MSSA) was used, comprising of two stages as decomposition and reconstruction. This technique helped the researchers in detecting known as decomposition and reconstruction. This technique helped the researchers in de- the FHR and separating the unwanted noise both in stationary and non-stationary signals. tecting the FHR and separating the unwanted noise both in stationary and non-stationary The main advantage of FECG monitoring is that it can be used remotely, in a non-clinical signals. The main advantage of FECG monitoring is that it can be used remotely, in a non- environment with high simplicity [7,80]. clinical environment with high simplicity [7,80]. 4. Importance of Electrode Configurations for FECG Measurement 4.4.1. Importance Electrode Configurations of Electrode Configurations for FECG Measurement 4.1. ElectrodeTable1 shows Configurations the comparison of electrode placement configurations over the maternal abdomenTable to1 shows monitor the the comparison FECG non-invasively. of electrode placement Comparisons configurations have been over made the based mater- on nalthe abdomen number of to electrodes monitor the placed FECG on thenon-invasi maternalvely. abdomen, Comparisons and the have filter been parameter made designbased onused the for number pre-processing, of electrodes such placed as bandwidth on the maternal (BW), sampling abdomen, frequency and the (Fs), filter resolution parameter (R) designof the A/Dused converter,for pre-processing, and the gain such (G). as bandwidth (BW), sampling frequency (Fs), resolution (R) of the A/D converter, and the gain (G).

SignalsSignals 2021 2021, 2, 2FOR FOR PEER PEER REVIEW REVIEW 9 9

Signals 2021, 2 400

TableTable 1. 1. Comparison Comparison of of different different electrode electrode placement placement for for FECG FECG monitoring. monitoring. Table 1. Comparison of different electrode placement for FECG monitoring. NumberNumber of of ProposedProposed NumberNumber of of Parameters’Parameters’ Specifica- Specifica- NumberParticipantsParticipants of and and NumberNumber of of Ges- Ges- MethodMethod for for No.No. AuthorAuthor ElectrodesNumberElectrodes of ElectrodesElectrodes Placement Placement AccuracyAccuracy Number of Proposed Method Parameters’tiontion ParticipantsTimeTime of of Data Data and Record- Record- tationtation Weeks Weeks ExtractionExtraction of of No. Author ElectrodesUsed Electrodes Placement Accuracy Gestation for Extraction of Used Speciﬁcation Time of Data Used inging ( s()s ) Weeks FECG FECGFECG Recording (s) CancellationCancellation of of SuccessSuccessSuccess rates forrates rates for for de- de- Cancellationmaternalmaternal of tem- tem- EquilateralEquilateral triangle triangle formation formation of of detectingtectingtecting the the P, the P, P, QRS, QRS, Chia et al., Equilateral triangle formation of three ParticipantsParticipants = 100 = 100 maternalplate template in 1st and ChiaChia et et al., al., Participants = 100QRS, and T waves plate in 1st and 1. 1.1. 3 3 threeelectrodesthree electrodes electrodes on the onabdomen on the the abdomen abdomen of pregnant of of N/AN/AN/A Recording = andand T T waves waves were were>18 weeks>18>18 weeks weeksin 1st and 2nd 200520052005 [81 [81] ][81] RecordingRecording = =10 10 min weremin 74.6%, 91.0%, 2nd2nd derivative derivative pregnantpregnantwomen. women. women. 10 min 74.6%,74.6%, 91.0%, 91.0%, and and derivative of and 79.3%, ofof abdominal abdominal 79.3%,79.3%, respectively respectively abdominal signals respectively signalssignals

BergveldBergveldBergveld et al., et et BandwidthBandwidthBandwidth = = 0.2 0.2 to toto 120 120 BetweenBetween 20Between and 20 20 and and 2. 2.2. 4 4 ParticipantsParticipantsParticipants = 37 = =37 37 ------al.,1986al., 1986 1986 [82] [82] [82] 120HzHz Hz 37 weeks3737 weeks weeks

SamplingSamplingSampling Frequency FrequencyFrequency = = AlgunaidiAlgunaidi et et ParticipantsParticipantsParticipants = 30 = =30 30 BetweenBetween 36Between and 36 36 Peakand and detection PeakPeak detection detection 3. 3.3. 4 4 =256 256256 Hz HzHz - - - et al.,al., 20112011 2011 [ 83[83] [83]] RecordingRecordingRecording = 60 s= =60 60 s s 38 weeks3838 weeks weeks Algorithm AlgorithmAlgorithm ResolutionResolutionResolution == 12=12 12 bitsbits bits

Signals 2021, 2 401

Table 1. Cont.

Number of Number of Number of Proposed Method Signals 2021, 2 FOR PEER REVIEW Parameters’ Participants and 10 No. Signals 2021 Author, 2 FOR PEER ElectrodesREVIEW Electrodes Placement Accuracy Gestation for Extraction of 10 Speciﬁcation Time of Data Used Weeks FECG Recording (s) Two scenarios were Two scenarios were Two scenariosused: 3-stage were used:used: 3-stage • 8 Participants 3-stage method:method: time 8 Participants• 8 Participants and method: time SamplingSampling Frequency Frequency = time frequency Sampling Frequency =recording and recording of 60 s of 60 s frequency KarvounisKarvounis et = 300300 Hz Hz and recording of 60 s analysis, complexfrequency 4. 4. Karvounis et 4 4 300 Hz was performed.was performed. 97.47%97.47% 20–41 weeks20–41 weeks analysis, com- 4.et al., 2007 [84] 4 Resolution = 12 bits was performed. 97.47% 20–41 weekswavelet, analysis, and com- al., 2007 [84] Resolution = 12 bits 5 Participants• and al., 2007 [84] ResolutionGain = 7800 = 12 bits • 5 Participants Heuristicplex wavelet, Gain = 7800 recording 5 Participants of plex wavelet, Gain = 7800 and recording of 15 algorithmand Heuristic 15 minand each recording was of 15 and Heuristic min each was per- algorithm performed.min each was per- algorithm formed. formed.

Fetal heart Fetal heart rates (FHR) ex- Fetal heartrates rates (FHR) ex- Emad. A. Ib- (FHR) extractedtracted from 5. Emad.Emad. A. A. Ib- 4 N/A N/A 60% 34 weeks tracted from 5. 5. rahim [15]. 4 4 N/AN/A N/AN/A 60%60% 34 weeks 34 weeksfrom fetalfetal phono- phonocar- Ibrahimrahim [15 [15].]. fetal phonocar- cardiographydiography (FPCG)diography (FPCG) (FPCG)

Bandwidth = 1–70 Hz Participants = 150 Graatsma et BandwidthBandwidth = 1–70 = Hz Participants = 150 Between 15 and Graatsma et Participants = 150 Between 15 and 6. Graatsma 5 - Sampling1–70 Hzfrequency = Time of recording = - Between 15 and - 6. 6. al., 2008 [85] 55 -- Sampling frequency =Time Time of recording of recording = - - 24 weeks - - et al.,al., 2008 2008 [85 [85]] Sampling300 frequency Hz 15 h 24 weeks 24 weeks 300 Hz = 15 h 15 h Rooijakkers = 300 Hz Participants = 5 Rooijakkers Participants = 5 7. et al., 2014 6 - - ParticipantsRecording = 5 = 20 min - 39 weeks - 7. Rooijakkerset al., 2014 6 - - Recording = 20 min - 39 weeks - 7. 6 - - Recording = - 39 weeks - et al., 2014[86] [86 ] each [86] 20 min eacheach

Signals 2021, 2 402

Table 1. Cont.

Number of Signals 2021, 2 FOR PEER NumberREVIEW of Number of Proposed Method 11 Signals 2021, 2 FOR PEER REVIEW Parameters’ Participants and 11 No. Author Electrodes Electrodes Placement Accuracy Gestation for Extraction of Speciﬁcation Time of Data Used Weeks FECG Recording (s)

WeightedWeighted Av-Av- Weighted Participants = 7 eragingeraging ofof VullingsVullings et al., et SamplingSampling frequency frequency = 1 Participants = 7 Averaging of 8. 8. Vullings et 88 Sampling frequency = 1 RecordingParticipants = = 7 91%91% 37–41 weeks 37–41 weeks MECG seg- 8. 2009 [87] 8 = 1 kHz Gain = 500 91% 37–41 weeksMECG segmentsMECG seg- al.,al., 20092009 [87][87] kHzkHz GainGain == 500500 RecordingRecording10 min == 1010 minmin (WAMES)mentsments (WAMES)(WAMES)

RooijakkersRooijakkers Rooijakkers SamplingSamplingSampling Frequency Frequency == 11 40 weeks during4040 weeksweeks duringduringPeak detection PeakPeak detectiondetection 9. 9.9. etet al.,al., 20122012 888 --- 91% 91%91% et al., 2012 [15] = 1kHz kHz labor laborlabor algorithmalgorithmalgorithm [15][15]

Andreotti et Participants = 10 Andreotti et ParticipantsParticipants = 10 = 10 10.10. Andreotti 88 -- - - 90%90% 20–2820–28 weeksweeks == 10. al.,al., 20142014 [88][88] 8 - - RecordingRecordingRecording = == 2020 minmin 90% 20–28 weeks - et al., 2014 [88] 20 min AdaptiveAdaptive R-R- ZhangZhang etet al.,al., ParticipantsParticipants == 7878 11.11. 1010 -- - - Participants = 78 -- 3rd3rd trimestertrimesterAdaptive wavewave R-wave detectiondetection Zhang2014 et al.,[68] Recording = 24 s each 11. 2014 [68] 10 - - RecordingRecording = = 24 s each - 3rd trimester detectionalgorithm 2014 [68] algorithm 24 s each algorithm

MartensMartens etet SequentialSequential esti-esti- 12.12. 1313 GainGain == 2020 - - 85% 85% 9 9 weeksweeks toto laborlabor al.,al., 20072007 [89][89] mationmation methodmethod

Signals 2021, 2 FOR PEER REVIEW 11

Weighted Av- eraging of Vullings et Sampling frequency = 1 Participants = 7 8. 8 91% 37–41 weeks MECG seg- al., 2009 [87] kHz Gain = 500 Recording = 10 min ments (WAMES)

Rooijakkers Sampling Frequency = 1 40 weeks during Peak detection 9. et al., 2012 8 - 91% kHz labor algorithm [15] Signals 2021, 2 403

Andreotti et Table 1. Cont. Participants = 10 10. 8 - - 90% 20–28 weeks = al., 2014 [88] NumberRecording of = 20 min Number of Number of Proposed Method Parameters’ Participants and Adaptive R- No. AuthorZhang et al., Electrodes Electrodes Placement Participants = 78 Accuracy Gestation for Extraction of 11. 10 - Speciﬁcation - Time of Data - 3rd trimester wave detection 2014 [68] Used Recording = 24 s each Weeks FECG Recording (s) algorithm

MartensMartens et al., et SequentialSequential esti- 12. 12. 1313 GainGain = 20= 20 - - 85% 85% 9 weeks to 9 labor weeks to labor 2007al., 2007 [89] [89] estimationmation method method Signals 2021, 2 FOR PEER REVIEW 12

Linear regression Taylor et al., 12–16 Sampling Frequency Participants = 241 Between 15 and Linear regres- 13. Taylor et al., 12–16 elec- Sampling Frequency = Participants = 241 85% Between 15to analyzeand QRS 13. 2003 [90] electrodes = 512 Hz Recording = 250 85% 33 weeks sion to analyze 2003 [90] trodes 512 Hz Recording = 250 33 weeks intervals QRS intervals

Oostendorp Homogeneous Sampling Frequency = Participants = 6 Between 20 and 14. et al.,1989 32 - volume con- 500 Hz Recordings = 37 40 weeks [91] duction model

Clifford et Sampling Frequency = 1 Between 35 and 15. 32 - - 91.2% - al., 2011 [92] kHz 41 weeks

Signals 2021, 2 FOR PEER REVIEW 12

Signals 2021, 2 404

Linear regres- Taylor et al., 12–16 elec- Sampling Frequency = Participants = 241 Between 15 and 13. 85% sion to analyze 2003 [90] trodes 512 Hz Recording = 250 33 weeks Table 1. Cont. QRS intervals Number of Number of Number of Proposed Method Parameters’ Participants and No. Author Electrodes Electrodes Placement Accuracy Gestation for Extraction of Speciﬁcation Time of Data Used Weeks FECG Recording (s)

Oostendorp HomogeneousHomogeneous Oostendorp SamplingSampling FrequencyFrequency = ParticipantsParticipants = 6 = 6 Between 20Between and 20 and 14. 14. et al.,1989 3232 - - volumevolume con- et al.,1989 [91] =500 HzHz RecordingsRecordings = 37 = 37 40 weeks 40 weeks [91] conductionduction model model

CliffordClifford et al., et SamplingSampling Frequency Frequency = 1 Between 35Between and 35 and 15. 15. 3232 - - -- 91.2% 91.2% - - 2011al., 2011 [92] [92] = 1kHz kHz 41 weeks 41 weeks

Signals 2021, 2 405

In Table1, the comparison of different electrode placements is listed, which can further be divided into three different ranges, as listed in Table2[93].

Table 2. Different groups of electrode placement for NIFECG.

No. of Electrodes Discussion Reference Advantages: For NIFECG monitoring, the electrodes are placed on the abdomen of pregnant women. Hence, less are used for fetus monitoring, resulting in a simple Group 1: range of electrodes procedure without any discomfort to pregnant women. [15,81–89] between 1 and 4 Disadvantages: The data recorded does not provide detailed information about the growth of the fetus. Advantages: It is evidenced that the more electrodes placed on the abdomen of Group 2, comprises a range pregnant women, the more comprehensive information can be gained about the between 10 and fetus, thus helping in analyzing the well-being of the fetus efﬁciently and effectively. [89,90] 20 electrodes. Disadvantages: Discomfort among the pregnant women due to several electrode placements on the abdomen of pregnant women. Advantages: Provides the detailed information about the growth of the fetus. Disadvantages: The deployment of 20 or more electrodes for FHR monitoring Group 3, comprises a range involves a complex setup procedure in addition to being expensive. [91,92] of more than 20 electrodes Furthermore, results in skin irritation and severe discomfort to pregnant women due to gel used in setting up the electrodes to establish effective electrode conﬁguration.

In terms of detection accuracy of FHR, it was noticed from Table1 that 97.47% was achieved with a four-electrode configuration through a three-step method, including time frequency analysis, complex wavelet, and Heuristic algorithm, at a sampling frequency of 300 Hz, resolution of 12 bits, with a gain of 7800. An accuracy of 91% was obtained with 8-electrode and 32-electrode configurations at a sampling frequency of 1 KHz. Addition- ally, from Table1, comparing the sequential method and liner regression method using 12–16 electrodes, an accuracy of 85% was achieved in FHR monitoring. The accuracy of the method used for NIFECG monitoring to determine the fetal heart rate is entirely dependent on the number of electrodes used for monitoring, placement of electrodes, gestation week, and parameters such as sampling frequency, resolution bits, and the gain. Furthermore, in a long-term monitoring, use of the conventional hydrogel electrodes would also significantly affect the accuracy since the hydrogel electrodes would dry out over time, thus affecting skin-electrode impedance.

4.2. Electrode Types In FHR monitoring, the hydrogel electrodes improve ion conductivity and provide impedance matching between the skin and electrodes, thus minimizing noises and improving signal quality. However, hydrogel electrodes degrade and collect hair and dirt in addition to the possibility of causing skin irritation whilst being used, so they must be frequently replaced [94–97]. Moreover, the hydrogel layer on the electrodes tends to dry out over time so the long-term monitoring capability to monitor fetal heart rate will be compromised [98,99]. In view of the shortcomings of hydrogel electrodes in FHR monitoring, the smart- based electrodes have attracted more attention [99–101]. Smart textiles exploit the universal- ity of textiles which cover all types of woven, nonwoven, and knitted materials/fabrics. In recent years, the interest in smart textiles as tools to continuously monitor physiological signals, such as ECG, has increased due to the high comfort and portability, unlike traditional monitoring systems using hydrogel electrodes which could create skin irritation [101,102]. Conventional textile manufacturing methods have been used to fabricate e-textile electrodes. An example shown in Figure 10a is the smart textile electrodes made through sewing conductive yarn to form electrodes. These electrodes are highly stretchable and ﬂexible due to the nature of interlocking stitching patterns. Gold- or silver-coated yarn [103] as well as stainless yarn [104] are commonly used due to their high conductivity and the Signals 2021, 2 FOR PEER REVIEW 14

4.2. Electrode Types In FHR monitoring, the hydrogel electrodes improve ion conductivity and provide impedance matching between the skin and electrodes, thus minimizing noises and improving signal quality. However, hydrogel electrodes degrade and collect hair and dirt in addition to the possibility of causing skin irritation whilst being used, so they must be frequently replaced [94–97]. Moreover, the hydrogel layer on the electrodes tends to dry out over time so the long-term monitoring capability to monitor fetal heart rate will be compromised [98,99]. In view of the shortcomings of hydrogel electrodes in FHR monitoring, the smart- based electrodes have attracted more attention [99–101]. Smart textiles exploit the univer- sality of textiles which cover all types of woven, nonwoven, and knitted materials/fabrics. In recent years, the interest in smart textiles as tools to continuously monitor physiological signals, such as ECG, has increased due to the high comfort and portability, unlike traditional monitoring systems using hydrogel electrodes which could create skin irritation [101,102]. Conventional textile manufacturing methods have been used to fabricate e-textile electrodes. An example shown in Figure 10a is the smart textile electrodes made through Signals 2021, 2 sewing conductive yarn to form electrodes. These electrodes are highly stretchable and 406 flexible due to the nature of interlocking stitching patterns. Gold- or silver-coated yarn [103] as well as stainless yarn [104] are commonly used due to their high conductivity and the fact that they are readily available in the market [39]. However, the drawback is to fact that they are readily available in the market [39]. However, the drawback is to build up build aup ﬁrm a firm contact contact between between skin andskin electrode and electrode surface surface without without applying applying too much too compression much compressionforce through force through a strap. a strap.

(a) (b)

FigureFigure 10. (a) 10.Sewn(a) ECG Sewn electrode; ECG electrode; (b) screen (b) screenprinted printed ECG electrode. ECG electrode. Reprinted (Reprinted with permission with permission from reffrom [39] ref Hasan, [39] Hasan, Muhammad Muhammad (2009)). (2009)).

In contrast,In contrast, printing printing techniques techniques (e.g., scr (e.g.,een printing) screen printing) have also have been also adapted been to adapted fab- to ricatefabricate electrodes electrodes on textiles. on textiles.The carbon-bas The carbon-baseded silicone silicone is commonly is commonly used as used an asinterface an interface betweenbetween skin and skin conductive and conductive layer layer to enhance to enhance the theskin skin impedance impedance matching matching and and to topro- provide vide aa solid/stable contact, contact, thus thus improving improving the ECG signal qualityquality andand minimizingminimizing thethenoise. noise.An An example example of of screen-printed screen-printed ECG ECG electrodes electrodes is shown is shown in Figure in Figure 10b, 10b, where where carbon-loaded carbon-loadedsilicone silicone is printed is printed directly directly onto textile onto wheretextile itwhere is needed it is needed to form to the form electrodes. the elec- These trodes.printed These electrodesprinted electrodes provide provide a higher a degreehigher ofdegree ﬂexibility of flexibility in terms in ofterms mass of production, mass production,customization, customization, impedance impedance matching, matching and noise, and noise reduction. reduction. Vojtech Vojtech et al. et developed al. de- a velopedthree-electrode a three-electrode ECG monitoringECG monitoring system syst usingem prefabricatedusing prefabricated conductive conductive textile [textile39,105 ,106]. [39,105,106].In comparison In comparison to conductive to conductive textile ortextile knitting, or knitting, sewing sewing enable enable enhanced enhanced freedom free- because dom becauseof the broad of the range broad of patternsrange of that patterns can be th usedat can and be the used ability and to the overlap ability stitches to overlap to create a stitchesdenser to create patch a ofdenser conductivity patch of within conductivity the fabric within [107]. the Kannaian fabric [107]. et al. deployedKannaian sewing et al. and embroidery to integrate silver-coated threads into fabrics to achieve electrodes for a textile- based ECG monitoring system [108]. There are several methods available to be employed to adhere the conductive substance to the chosen substrate: dip and dry, sputtering, screen printing, chemical etching, stencil printing, and permeation [106,107]. The electrode type and method deployed for ECG monitoring depend on the properties of the conductive substance and the substrate. However, the screen printing can produce higher printing volume with fewer printing cycles to achieve a stable conductive pattern on e-textile surfaces and is also a standard mass production method used in the textile industry.

4.3. Current ECG Monitoring Development with Smart Textile Electrodes Traditionally, ECG monitoring relies upon adhesive hydrogel electrodes. There are some key challenges that arise with this type of electrode including discomfort to patients due to skin irritation caused by hydrogels [109,110], difﬁcult positioning when more electrodes are needed, and less accurate signal quality over longer period due to the degradation of hydrogel [105]. Over the past few years, novel concepts of using smart textile electrodes for monitoring ECG electrodes have emerged. Smart textiles-based electrodes are preferred for ECG monitoring as they can be deployed for longer term monitoring easily without any risk of hydrogel degradation and provide enhanced comfort to patients due to no skin irritation [111–113]. Some examples of current development in smart textile ECG electrodes are summarized in Table3. Signals 2021, 2 407

Table 3. Some examples of smart textile ECG research.

No. Method Material System Integration Reference 1. Knitting Stainless steel T-shirt [113] Knitting and Stainless steel 2. T-shirt [114] Embroidery ﬁlament, nylon fabric 3. Weaving and Knitting Silver yarns Chest band [115] 4. Screen printing Silver ink Chest band [116] 5. Electroless platting Silver chloride Smart garment [117]

With the increase in cardiovascular diseases, there has been a growing interest in developing wearable devices, speciﬁcally smart textiles for ECG monitoring, that can continuously monitor cardiac activity without causing discomfort to patients [105]. A smart textile-based T-shirt achieved by knitting and embroidering stainless steel yarns on nylon fabric was deployed for long-term ECG monitoring, thus providing no skin irritation to patients, and enhancing the comfort to patients [113,114]. The authors of [115,116] used the woven and screen-printed smart textile electrodes for ECG monitoring. The authors of [117] achieved a silver chloride electrode using electroless plating for the ECG monitoring. However, the use of smart textile electrodes in FECG and NIFECG has yet to be reported due to the following barriers: (a) the optimized number of electrodes required for fetal heart rate monitoring needs to be investigated, and (b) the signal quality of FECG needs to be improved due to the small amplitude compared with MECG. The deployment of smart textile electrodes in non-invasive fetal electrocardiography (NIFECG) for fetal heart rate (FHR) monitoring will revolutionize the healthcare monitoring in a way that smart textile-based electrodes would provide enhanced comfort to patients and no skin irritation during long-term monitoring.

5. Conclusions This paper provided a review of monitoring techniques including CTG, PPG, and Doppler sound, with the emphasis on fetal electrocardiogram (FECG), to detect fetal heart rate to determine the well-being and growth of the fetus. The importance of non- invasive FECG in contrast to invasive FECG in terms of comfort to patients was mentioned and highlighted in this paper. The different electrode conﬁgurations used in NIFECG monitoring were also discussed. It was shown that NIFFECG with fewer electrodes is more suitable for home use environments because the setup requires minimum preparation time, providing the essential FECG information and enhancing the comfort to the patients. One of the key drawbacks of current NIFECG is the use of hydrogel electrodes, which tend to degrade over time. The positioning of these electrodes may also increase the complexity for being used at home by pregnant women. These drawbacks could be overcome by the introduction of smart textile-based electrodes, which are realized using either conventional textile manufacturing methods or printing. These electrodes do not need the assistance of a hydrogel layer to provide adhesion. Instead, the garment on which the electrodes are formed provides contraction to attach the electrodes against the skin. In addition, these electrodes are realized on the correct position and therefore the complex setup process is minimized. It is suggested that future research could focus on improving the contact between smart textile electrodes and skin. This is of importance since the quality of the FECG signal is an indicator of the well-being of the fetus to diagnose any chronic diseases at an early stage of pregnancy, and thus to help in taking the essential preventive and precautionary steps on time. Moreover, electrode number and positioning need to be investigated in the context of smart textiles to provide a compromise between wearability, useability, washability, and signal quality, which is a future research direction in NIFECG monitoring using smart textile electrodes. Therefore, the deployment of smart Signals 2021, 2 408

textile electrodes in NIFECG monitoring will play a crucial role in long-term monitoring and comfort to patients. It is envisioned that the future perspective of smart textiles NIFECG in medical diagnosis will not be limited for the personal use because the device is believed to beneﬁt the patients by carrying out the measurement either at hospitals by doctors or at home by themselves. There has been an ever-increasing utilization of cloud-based computing and Internet of things (IoT). Smart textile NIFECG should be light-weight and low-cost, providing data security and enhancing data transmission rate. Through cloud computing, the smart textile-based NIFECG measurement could be wirelessly connected to other medical IoT devices or a centralized hub, enabling accessing the essential health information from different locations.

Author Contributions: G.A. wrote the paper; Y.W. helped with revision; Both G.A. and Y.W. contributed to editing the paper. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by Nottingham Trent University strategic research theme, Medical Technologies and Advanced Materials. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Not applicable. Conﬂicts of Interest: The authors declare no conﬂict of interest.

References

1. Ungureanu, G.; Gussi, I.; Wolf, W.; T, arălungă, D.; Pasca, S.; Strungaru, R. Prenatal Telemedicine Advances in Fetal Monitoring. 2011. Available online: https://scholar.google.com.hk/scholar?hl=en&as_sdt=0%2C5&q=Ungureanu%2C+Georgeta+%26+ Gussi%2C+Ilinca+%26+Wolf%2C+Werner+%26+%C8%9Aar%C4%83lung%C4%83%2C+Drago%C8%99+%26+Pasca%2C+ Sever+%26+Strungaru%2C+Rodica.+%282011%29.+Prenatal+Telemedicine+Advances+in+Fetal+Monitoring.+10.5772%2F135 33.&btnG= (accessed on 23 June 2021). [CrossRef] 2. Hasan, M.A.; Reaz, M.B.I.; Ibrahimy, M.I.; Hussain, M.S.; Uddin, J. Detection and processing techniques of FECG signal for fetal monitoring. Biol. Proced. Online 2009, 11, 263–295. [CrossRef] 3. Chez, B.F.; Baird, S.M. Electronic Fetal Heart Rate Monitoring: Where Are We Now? J. Perinat. Neonatal Nurs. 2011, 25, 180–192. [CrossRef] 4. Neilson, J.P. Fetal electrocardiogram (ECG) for fetal monitoring during labour. Cochrane Database Syst. Rev. 2006, 3.[CrossRef] 5. Freeman, R.K.; Garite, T.J.; Nageotte, M.P.; Miller, L.A. Fetal Heart Rate Monitoring; Lippincott Williams & Wilkins: Philadelphia, PA, USA, 2012. 6. Peters, C.H.; ten Broeke, E.D.; Andriessen, P.; Vermeulen, B.; Berendsen, R.C.; Wijn, P.F.; Oei, S.G. Beat-to-beat detection of fetal heart rate: Doppler ultrasound cardiotocography compared to direct ECG cardiotocography in time and frequency domain. Physiol. Meas. 2004, 25, 585. [CrossRef][PubMed] 7. Vullings, R.; Peters, C.; Mischi, M.; Oei, G.; Bergmans, J. Maternal ECG removal from non-invasive fetal ECG recordings. In Proceedings of the EMBS’06. 28th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, New York, NY, USA, 30 August–3 September 2006; pp. 1394–1397. 8. Hon, E.H. Apparatus for continuous monitoring of the fetal heart rate. Yale J. Biol. Med. 1960, 32, 397. 9. Fetal Heart Monitoring. Journal of Obstetric, Gynecologic & Neonatal Nursing. 2018. Available online: https://www.jognn.org/ article/S0884-2175(18)30322-8/fulltext (accessed on 23 June 2021). [CrossRef] 10. Jezewski, J.; Roj, D.; Wrobel, J.; Horoba, K. A novel technique for fetal heart rate estimation from Doppler ultrasound signal. Biomed. Eng. Online 2011, 10.[CrossRef] 11. Black, R.S.; Campbell, S. Cardiotocography versus Doppler. Ultrasound Obstet. Gynecol. 1997, 9, 148–151. [CrossRef][PubMed] 12. Peters, M.; Crowe, J.; Piéri, J.F.; Quartero, H.; Hayes-Gill, B.; James, D.; Shakespeare, S. Monitoring the fetal heart non-invasively: A review of methods. J. Perinat. Med. 2001, 29, 408–416. [CrossRef][PubMed] 13. Strasburger, J.F.; Cheulkar, B.; Wakai, R.T. Magnetocardiography for fetal arrhythmias. Heart Rhythm Off. J. Heart Rhythm Soc. 2008, 5, 1073. [CrossRef] 14. Adithya, P.C.; Sankar, R.; Moreno, W.A.; Hart, S. Trends in fetal monitoring through phonocardiography: Challenges and future directions. Biomed. Signal Process. Control 2017, 33, 289–305. [CrossRef] 15. Andreotti, F.; Riedl, M.; Himmelsbach, T.; Wedekind, D.; Wessel, N.; Stepan, H.; Schmieder, C.; Jank, A.; Malberg, H.; Zaunseder, S. Robust fetal ECG extraction and detection from abdominal leads. Physiol. Meas. 2014, 35, 1551–1568. [CrossRef] Signals 2021, 2 409

16. Assaleh, K. Extraction of fetal electrocardiogram using adaptive neuro-fuzzy interference systems. IEEE Trans. Biomed. Eng. 2007, 54, 59–68. [CrossRef] 17. Behar, J.; Johnson, A.; Clifford, G.D.; Oster, J. A comparison of single channel fetal ECG extraction methods. Ann. Biomed. Eng. 2014, 42, 1340–1353. [CrossRef][PubMed] 18. Gao, X. On the improved correlative prediction scheme for aliased electrocardiogram (ECG) data compression. In Proceedings of the 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), San Diego, CA, USA, 28 August–1 September 2012; pp. 6180–6183. 19. Ghodsi, M.; Hassani, H.; Sanei, S. Extracting fetal heart signal from noisy maternal ECG by singular spectrum analysis. J. Stat. Interface Spec. Issue Appl. SSA 2010, 3, 399–411. [CrossRef] 20. Hyvarinen, A. Fast and robust fixed-point algorithms for independent component analysis. IEEE Trans. Neural Netw. 1999, 10, 626–634. [CrossRef] 21. Immanuel, J.J.R.; Prabhu, V.; Christopheraj, V.J.; Sugumar, D.; Vanathi, P.T. Separation of maternal and fetal ECG signals from the mixed source signal using FASTICA. Procedia Eng. 2012, 30, 356–363. [CrossRef] 22. Jafari, M.G.; Chambers, J.A. Fetal electrocardiogram extraction by sequential source separation in the wavelet domain. IEEE Trans. Biomed. Eng. 2005, 52, 390–400. [CrossRef] 23. Kropfl, M.; Modre-Osprian, R.; Schreier, G.; Hayn, D. A robust algorithm for fetal QRS detection using non-invasive maternal abdominal ECGs. Comput. Cardiol. 2013, 40, 313–316. 24. Kumar, P.; Sharma, S.K.; Prasad, S. Detection of FECG from multivariate abdominal recordings using wavelets and neuro-fuzzy systems. Int. J. Eng. Adv. Technol. Stud. 2013, 2, 45–51. 25. Melillo, P.; Santoro, D.; Vadursi, M. Detection and compensation of inter-channel time offsets in indirect fetal ECG sensing. IEEE Sens. J. 2014, 14, 2327–2334. [CrossRef] 26. Panigrahy, D.; Sahu, P.K. Extraction of fetal electrocardiogram (ECG) by extended state Kalman filtering and adaptive neuro-fuzzy inference system (ANFIS) based on single channel abdominal recording. Sadhana 2015, 40, 1091–1104. [CrossRef] 27. Poian, G.D.; Bernardini, R.; Rinaldo, R. Separation and analysis of fetal-ECG signals from compressed sensed abdominal ECG recordings. IEEE Trans. Biomed. Eng. 2016, 63, 1269–1279. [CrossRef][PubMed] 28. Reza, S.; Shamsollahi, M.B.; Jutten, C.; Clifford, G.D. A nonlinear Bayesian filtering framework for ECG denoising. IEEE Trans. Biomed. Eng. 2007, 54, 2172–2185. 29. Rooijakkers, M.J.; Rabotti, C.; de Lau, H.; Oei, S.G.; Bergmans, J.W.M.; Mischi, M. Feasibility study of a new method for low-complexity fetal movement detection from abdominal ECG recordings. IEEE J. Biomed. Health Inform. 2016, 20, 1361–1368. [CrossRef] 30. Selvaraj, R.; Kanagaraj, B. A multi-stage adaptive singular value decomposition approach for fetal ECG signal extraction in multichannel input system for prenatal health monitoring. Asian J. Inf. Technol. 2016, 15, 1049–1055. 31. Sameni, R.; Clifford, G.D. A review of fetal ECG signal processing; issues and promising directions. Open Pacing Electrophysiol. Ther. J. 2010, 3, 4. [CrossRef] 32. von Steinburg, S.P.; Boulesteix, A.; Lederer, C.; Grunow, S.; Schiermeier, S.; Hatzmann, W.; Daumer, M. What is the “normal” fetal heart rate. PeerJ 2013.[CrossRef] 33. Silva, I.; Behar, J.; Sameni, R.; Zhu, T.-T.; Oster, J.; Clifford, G.D.; Moody, G.B. Noninvasive fetal ECG: The PhysioNet/computing in cardiology challenge 2013. In Proceedings of the IEEE Conference of Computing in Cardiology (CinC), Zaragoza, Spain, 22–25 September 2013; pp. 149–152. 34. Song, S.; Rooijakkers, M.J.; Harpe, P.; Rabotti, C.; Mischi, M.; van Roermund, A.H.M.; Cantatore, E. A noise reconfigurable current-reuse resistive feedback amplifier with signal-dependent power consumption for fetal ECG monitoring. IEEE Sens. J. 2016, 16, 8304–8313. [CrossRef] 35. Tadi, M.J.; Lehtonen, E.; Hurnanen, T.; Koskinen, J.; Eriksson, J.; Pänkäälä, M.; Teräs, M.; Koivisto, T. A real-time approach for heart rate monitoring using a Hilbert transform in seismocardiograms. Physiol. Meas. 2016, 37, 1885–1909. [CrossRef] 36. Yeh, H.-M.; Chang, Y.-C.; Lin, C.; Yeh, C.-H.; Lee, C.-N.; Shyu, M.-K.; Hung, M.-H.; Hsiao, P.-N.; Wang, Y.-H.; Tseng, Y.-H.; et al. A new method to derive fetal heart rate from maternal abdominal electrocardiogram monitoring fetal heart rate during cesarean section. PLoS ONE 2015, 10, e0117509. 37. Zheng, W.; Li, X.-L.; Wei, X.-Y.; Liu, H.-X. Foetal ECG extraction by support vector regression. Electron. Lett. 2016, 52, 506–507. 38. Adam, J. The Future of Fetal Monitoring. Rev. Obstet. Gynecol. 2012, 5, e132. 39. Hasan, M.; Ibrahimy, M.; Reaz, M. Fetal ECG Extraction from Maternal Abdominal ECG Using Neural Network. J. Softw. Eng. Appl. 2009, 2, 330–334. [CrossRef] 40. Graatsma, E.M.; Jacod, B.C.; Van Egmond, L.A.J.; Mulder, E.J.H.; Visser, G.H.A. Fetal electrocardiography: Feasibility of long-term fetal heart rate recordings. BJOG Int. J. Obstet. Gynaecol. 2009, 116, 334338. [CrossRef] 41. Graatsma, E.M. Monitoring of Fetal Heart Rate and Uterine Activity; Utrecht University: Utrecht, The Netherlands, 2010. 42. Karvounis, E.C.; Papaloukas, C.; Fotiadis, D.I.; Michalis, L.K. Fetal heart rate extraction from composite maternal ECG using complex continuous wavelet transform. In Computers in Cardiology; IEEE: Piscataway, NJ, USA, 2004; pp. 737–740. 43. Zhongliang, L.U.O. Fetal Electrocardiogram Extraction using Blind Source Separation and Empirical Mode Decomposition. J. Comput. Inf. Syst. 2012, 8, 4825–4833. Signals 2021, 2 410

44. Zheng, W.; Liu, H.; He, A.; Ning, X.; Cheng, J. Singlelead fetal electrocardiogram estimation by means of combining Rpeak detection, resampling and comb filter. Med. Eng. Phys. 2010, 32, 708–719. [CrossRef][PubMed] 45. Martínek, R.; Zidek, J. A System for Improving the Diagnostic Quality of Fetal Electrocardiogram. Prz. Elektrotech. 2012, 164–173. Available online: https://scholar.google.com.tw/scholar?hl=en&as_sdt=0%2C5&q=46.%09Mart%C3%ADnek%2C+R.%2C+ %26+Zidek%2C+J.+%282012%29.+A+System+for+Improving+the+Diagnostic+Quality+of+Fetal+Electrocardiogram.+Przegl% C4%85d+El-ektrotechniczny%2C+164-173.&btnG= (accessed on 23 June 2021). 46. Sree, T.H.; Garimella, M.; Bandari, A.; Patel, I. Microcontroller Based Fetal Heart Rate Monitoring using Intelligent Biosystem. In Proceedings of the 3rd International Conference on Electronics, Biomedical Engineering and Its Applications (ICEBEA’2013), Singapore, 29–30 April 2013. 47. Sun, Y.; Hu, S.; Azorin-Peris, V.; Greenwald, S.; Chambers, J.; Zhu, Y. Motion-compensated noncontact imaging photoplethysmography to monitor cardiorespiratory status during exercise. J. Biomed. Opt. 2011, 16, 077010. [CrossRef][PubMed] 48. Nitzan, M.; Khanokh, B.; Slovik, Y. The difference in pulse transit time to the toe and finger measured by photoplethysmography. Physiol. Meas. 2002, 23, 85. [CrossRef][PubMed] 49. Gan, K.B.; Ali, M.M.; Zahedi, E. Two Channel Abdominal PPG Instrumentation. In Proceedings of the 4th Kuala Lumpur International Conference on Biomedical Engineering 2008, Kuala Lumpur, Malaysia, 25–28 June 2008; Springer: Berlin/Heidelberg, Germany, 2008; pp. 691–693. 50. Gan, K.B.; Zahedi, E.; Ali, M.A.M. Transabdominal fetal heart rate detection using NIR photopleythysmography: Instrumentation and clinical results. IEEE Trans. Biomed. Eng. 2009, 56, 2075–2082. [PubMed] 51. Gan, K.B.; Zahedi, E.; Ali, M.M. Feasibility of Fetal Photoplethysmography Signal Extraction using Adaptive Noise Cancelling. In Proceedings of the 3rd Kuala Lumpur International Conference on Biomedical Engineering 2006, Kuala Lumpur, Malaysia, 11–14 December 2006; Springer: Berlin/Heidelberg, Germany, 2006; pp. 387–390. 52. Ahsan, K.; Emad, I.; Sayaka, O.; Yoshitaka, K. Validation of beat by beat fetal heart signals acquired from four-channel fetal phonocardiogram with fetal electrocardiogram in healthy late pregnancy. Sci. Rep. 2018, 8.[CrossRef] 53. Heart Rate. Available online: http://fabacademy.org/archives/2013/students/contonente.javier/week16/week16.html (accessed on 16 February 2021). 54. Cesarelli, M.; Romano, M.; Bifulco, P.; Fedele, F.; Bracale, M. An algorithm for the recovery of fetal heart rate series from CTG data. Comput. Biol. Med. 2007, 37, 663669. [CrossRef] 55. Williams, B.; Arulkumaran, S. Cardiotocography and medicolegal issues. Best Pract. Res. Clin. Obstet. Gynaecol. 2004, 18, 457–466. [CrossRef] 56. Ugwumadu, A. Understanding cardiotocographic patterns associated with intrapartum fetal hypoxia and neurologic injury. Best Pract. Res. Clin. Obstet. Gynaecol. 2013, 27, 509–536. [CrossRef] 57. Alfirevic, Z.; Devane, D.; Gyte, G.M. Continuous cardiotocography (CTG) as a form of electronic fetal monitoring (EFM) for fetal assessment during labour. Cochrane Database Syst. Rev. 2006, 3.[CrossRef] 58. Paul, H.; Rik, V.; Alexander, K.; Jan, B.; van Judith, L.; Piero, T.; Massimo, M. Doppler Ultrasound Technology for Fetal Heart Rate Monitoring: A Review. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 2019.[CrossRef] 59. Pregnancy and Antental Care. Available online: https://steemit.com/health/@doctorhealth/pregnancy-antenatal-care-and- counselling (accessed on 10 February 2021). 60. Voicu, I.; Ménigot, S.; Kouamé, D.; Girault, J. New Estimators and Guidelines for Better Use of Fetal Heart Rate Estimators with Doppler Ultrasound Devices. Comput. Math. Methods Med. 2014, 2014.[CrossRef][PubMed] 61. Karlsson, B.; Foulquière, K.; Kaluzynski, K.; Tranquart, F.; Fignon, A.; Pourcelot, D.; Berson, M.; Pourcelot, L. The DopFet system: A new ultrasonic Doppler system for monitoring and characterization of fetal movement. Ultrasound Med. Biol. 2000, 26, 1117–1124. [CrossRef] 62. Jezewski,˙ J.; Wróbel, J.; Horoba, K.; Cholewa, D.; Gacek, A.; Kupka, T.; Matonia, A. Monitoring of mechanical and electrical activity of fetal heart: The nature of signals. Arch. Perinat. Med. 2002, 8, 40–46. 63. Van Geijn, H.P.; Copray, F.J.A. A Critical Appraisal of Fetal Surveillance; Academic Hospital of the Free University: Amsterdam, The Netherlands, 1994. 64. Crowe, J.A.; Herbert, J.M.; Huang, X.B.; Reed, N.; Woolfson, M.S.; Rassi, D.; Zhuravlev, Y.E.; Emery, S.J. Sequential recording of the abdominal fetal electrocardiogram and magnetocardiogram. Physiol. Meas. 1995, 16, 43–47. [CrossRef] 65. Goodlin, R.C. History of fetal monitoring. Am. J. Obstet. Gynecol. 1979, 133, 323–352. [CrossRef] 66. Solum, T.; Ingemarsson, I.; Nygren, A. The accuracy of abdominal ECG for fetal electronic monitoring. J. Perinat. Med. 1980, 8, 142–149. [CrossRef] 67. Khamene, A.; Negahdaripour, S. A new method for the extraction of fetal ECG from the composite abdominal signal. Biomed. Eng. IEEE Trans. 2000, 47, 507–516. [CrossRef][PubMed] 68. Clifford, G.; Sameni, R.; Ward, J.; Robinson, J.; Wolfberg, A.J. Clinically accurate fetal ECG parameters acquired from maternal abdominal sensors. Am. J. Obstet. Gynecol. 2011, 205, 47.e1–47.e5. [CrossRef] 69. Kahankova, R.; Martinek, R.; Jaros, R.; Behbehani, K.; Matonia, A.; Jezewski, M.; Behar, J.A. A Review of Signal Processing Techniques for Non-Invasive Fetal Electrocardiography. IEEE Rev. Biomed. Eng. 2020, 13, 51–73. [CrossRef] 70. Monica Healthcare. Available online: http://www.monicahealthcare.com/ (accessed on 6 August 2019). 71. Mindchild. Available online: http://www.mindchild.com/ (accessed on 6 August 2019). Signals 2021, 2 411

72. Nemo Healthcare, Community Research and Development Information Service. Available online: https://nemohealthcare.com/en/ (accessed on 6 August 2019). 73. Cohen, W.R.; Ommani, S.; Hassan, S.; Mirza, F.G.; Solomon, M.; Brown, R.; Schifrin, B.S.; Himsworth, J.M.; Hayes-Gill, B.R. Accuracy and reliability of fetal heart rate monitoring using maternal abdominal surface electrodes. Acta Obstet. Gynecol. Scand. 2012, 91, 1306–1313. [CrossRef] 74. Hon, E.; Hess, O. The clinical value of fetal electrocardiography. Am. J. Obstet. Gynecol. 1960, 79, 1012–1023. [CrossRef] 75. Algunaidi, M.S.M.; Ali, M.M.; Gan, K.B.; Zahedi, E. Fetal heart rate monitoring based on adaptive noise cancellation and maternal QRS removal window. Eur. J. Sci. Res. 2009, 27, 565–575. 76. Goell, P.; Rai, S.; Chandra, M.; Gupta, V.K. Analysis of LMS Algorithm in Wavelet Domain. In Proceedings of the Conference on Advances in Communication and Control Systems Systems (CAC2S 2013), Dehradun, India, 6–8 April 2013. 77. Unser, M.; Aldroubi, A. A review of wavelets in biomedical applications. Proc. IEEE 1996, 84, 626–638. [CrossRef] 78. Abdulhay, E.W.; Oweis, R.J.; Alhaddad, A.M.; Sublaban, F.N.; Radwan, M.A.; Almasaeed, H.M. Non-Invasive Fetal Heart Rate Monitoring Techniques: Review article. Biomed. Sci. Eng. 2014, 2, 53–67. 79. Chia, E.L.; Ho, T.F.; Rauff, M.; Yip, W.C.L. Cardiac time intervals of normal fetuses using noninvasive fetal electrocardiography. Prenat. Diagn. 2005, 25, 546–552. [CrossRef][PubMed] 80. Algunaidi, M.S.; Ali, M.M.; Islam, M.F. Evaluation of an improved algorithm for fetal QRS detection. Int. J. Phys. Sci. 2011, 6, 213–220. 81. Karvounis, E.C.; Tsipouras, M.G.; Fotiadis, D.I.; Naka, K.K. An Automated Methodology for Fetal Heart Rate Extraction from the Abdominal Electrocardiogram. IEEE Trans. Inf. Technol. Biomed. 2007, 11, 628–638. [CrossRef][PubMed] 82. Rooijakkers, M.J.; Rabotti, C.; Oei, S.G.; Mischi, M. Low-complexity R-peak detection for ambulatory fetal monitoring. Physiol. Meas. 2012, 33, 1135–1150. [CrossRef][PubMed] 83. Rooijakkers, M.J.; Song, S.; Rabotti, C.; Oei, S.G.; Bergmans, J.W.M.; Cantatore, E.; Mischi, M. Influence of Electrode Placement on Signal Quality for Ambulatory Pregnancy Monitoring. Comput. Math. Methods Med. 2014, 2014, 960980. [CrossRef] 84. Vullings, R.; Peters, C.H.L.; Sluijter, R.J.; Mischi, M.; Oei, S.G.; Bergmans, J.W.M. Dynamic segmentation and linear prediction for maternal ECG removal in antenatal abdominal recordings. Physiol. Meas. 2009, 30, 291–307. [CrossRef][PubMed] 85. Jie-Min, Z.; Xiao-Lin, H.; Qun, G.; Tie-Bing, L.; Ping, L.; Ying, Z.; Hong-Xing, L. Some regularity on how to locate electrodes for higher fECG SNRs. Chin. Phys. B 2015, 24, 038702. 86. Martens, S.M.M.; Rabotti, C.; Mischi, M.; Sluijter, R.J. A robust fetal ECG detection method for abdominal recordings. Physiol. Meas. 2007, 28, 373–388. [CrossRef][PubMed] 87. Taylor, M.J.; Smith, M.J.; Thomas, M.; Green, A.R.; Cheng, F.; Oseku-Afful, S.; Wee, L.Y.; Fisk, N.M.; Gardiner, H.M. Non-invasive fetal electrocardiography in singleton and multiple pregnancies. BJOG Int. J. Obstet. Gynaecol. 2003, 110, 668–678. [CrossRef] 88. Oostendorp, T.F.; Van Oosterom, A.; Jongsma, H.W. The fetal ECG throughout the second half of gestation. Clin. Phys. Physiol. Meas. 1989, 10, 147–160. [CrossRef] 89. Marchon, N.; Naik, G. Electrode positioning for monitoring Fetal ECG: A review. In Proceedings of the 2015 International Conference on Information Processing (ICIP), Pune, India, 16–19 December 2015; pp. 5–10. [CrossRef] 90. Strazdiene,˙ E.; Blaževiˇc,P.; Vegys, A.; Dapkunien¯ e,˙ K. New Tendencies of Wearable Electronics Application in Smart Clothing. Electron. Electr. Eng. 2007, 73, 21–24. 91. Rodrigues, R. Fetal beat detection in abdominal ECG recordings: Global and time adaptive approaches. Physiol. Meas. 2014, 35, 1699–1711. [CrossRef][PubMed] 92. Panigrahy, D.; Rakshit, M.; Sahu, P.K. An efficient method for fetal ECG extraction from single channel abdominal ECG. In Proceedings of the 2015 International Conference on Industrial Instrumentation and Control (ICIC), Pune, India, 28–30 May 2015; pp. 1083–1088. [CrossRef] 93. Kennedy, R.G. Electronic fetal heart rate monitoring: Retrospective reflections on a twentieth-century technology. J. R. Soc. Med. 1998, 91, 244–250. [CrossRef][PubMed] 94. Ivanoska-Dacikj, A.; Stachewicz, U. Smart textiles and wearable technologies—Opportunities offered in the fight against pandemics in relation to current COVID-19 state. Rev. Adv. Mater. Sci. 2020, 59, 487–505. [CrossRef] 95. Hughes-Riley, T.; Dias, T.; Cork, C. A Historical Review of the Development of Electronic Textiles. Fibers 2018, 6, 34. [CrossRef] 96. Tsukada, Y.T.; Tokita, M.; Murata, H.; Hirasawa, Y.; Yodogawa, K.; Iwasaki, Y.-K.; Kasai, N.; Shimizu, W.; Nakashima, H.; Tsukada, S. Validation of wearable textile electrodes for ECG monitoring. Heart Vessels 2019, 34, 1203–1211. [CrossRef] 97. Paul, G.; Torah, R.; Beeby, S.; Tudor, J. A printed, dry electrode frank configuration vest for ambulatory vectorcardiographic monitoring. Smart Mater. Struct. 2016, 26, 025029. [CrossRef] 98. Komolafe, A.; Torah, R.; Wei, Y.; Nunes-Matos, H.; Li, M.; Hardy, D.; Dias, T.; Tudor, M.; Beeby, S. Integrating Flexible Filament Circuits for E-Textile Applications. Adv. Mater. Technol. 2019, 4, 1900176. [CrossRef] 99. Schwarz, A.; Hakuzimana, J.; Kaczynska, A.; Banaszczyk, J.; Westbroek, P.; McAdams, E.; Moody, G.; Chronis, Y.; Priniotakis, G.; De Mey, G.; et al. Gold coated para-aramid yarns through electroless deposition. Surf. Coat. Technol. 2010, 204, 1412–1418. [CrossRef] 100. Odhiambo, S.A.; De Mey, G.; Hertleer, C.; Schwarz, A.; van Langenhove, L. Discharge characteristics of poly(3,4-ethylene dioxythiophene): Poly(styrenesulfonate) (PEDOT:PSS) textile batteries; comparison of silver coated yarn electrode devices and pure stainless steel filament yarn electrode devices. Text. Res. J. 2014, 84, 347–354. [CrossRef] Signals 2021, 2 412

101. An, X.; Stylios, G.K. A Hybrid Textile Electrode for Electrocardiogram (ECG) Measurement and Motion Tracking. Materials 2018, 11, 1887. [CrossRef] 102. Achilli, A.; Pani, D.; Bonﬁglio, A. Characterization of Screen-Printed Textile Electrodes Based on Conductive Polymer for ECG Acquisition; IEEE: New York, NY, USA, 2017. [CrossRef] 103. Lee, Y.-D.; Chung, W.-Y. Wireless sensor network based wearable smart shirt for ubiquitous health and activity monitoring. Sens. Actuator B Chem. 2009, 140, 390–395. [CrossRef] 104. Acar, G.; Ozturk, O.; Golparvar, A.J.; Elboshra, T.A.; Böhringer, K.; Yapici, M.K. Wearable and Flexible Textile Electrodes for Biopotential Signal Monitoring: A review. Electronics 2019, 8, 479. [CrossRef] 105. Kannaian, T.; Neelaveni, R.; Thilagavathi, G. Design and development of embroidered textile electrodes for continuous measurement of electrocardiogram signals. J. Ind. Text. 2013, 42, 303–318. [CrossRef] 106. Alzaidi, A.; Zhang, L.; Bajwa, H. Smart textiles based wireless ECG system. In Proceedings of the 2012 IEEE Long Island Systems, Applications and Technology Conference (LISAT), Farmingdale, NY, USA, 4 May 2012; pp. 1–5. [CrossRef] 107. Smart Fabrics, a Technology That Revolutionizes Experiences—Ignasi Sayol. Available online: https://nemohealthcare.com/en/ (accessed on 13 June 2020). 108. Tang, D.H.; Gilligan, A.M.; Romero, K. Economic burden and disparities in healthcare resource use among adult patients with cardiac arrhythmia. Appl. Health Econ. Health Policy. 2014, 12, 59–71. [CrossRef] 109. Steinberg, C.; Bennett, M.T.; Krahn, A.D. Extended ECG Monitoring. In Cardiac Arrhythmias, Pacing and Sudden Death; Kowey, P., Piccini, J.P., Naccarelli, G., Eds.; Springer: Cham, Switzerland, 2017; pp. 48–59. 110. Priori, S.G. Survivors of out-of-hospital cardiac arrest with apparently normal heart. Need for deﬁnition and standardized clinical evaluation. Consensus Statement of the Joint Steering Committees of the Unexplained Cardiac Arrest Registry of Europe and of the Idiopathic Ventricular Fibrillation Registry of the United States. Circulation 1997, 95, 265–272. 111. Tomson, T.T.; Passman, R. Current and Emerging Uses of Insertable Cardiac Monitors: Evaluation of Syncope and Monitoring for Atrial Fibrillation. Cardiol. Rev. 2017, 25, 22–29. [CrossRef][PubMed] 112. Steinberg, C.; Philippon, F.; Sanchez, M.; Fortier-Poisson, P.; O’Hara, G.; Molin, F.; Sarrazin, J.F.; Nault, I.; Blier, L.; Roy, K.; et al. A Novel Wearable Device for Continuous Ambulatory ECG Recording: Proof of Concept and Assessment of Signal Quality. Biosensors 2019, 9, 17. [CrossRef] 113. Zheng, J.; Zhang, Z.; Wu, T.; Zhang, Y. A wearable mobihealth care system supporting real-time diagnosis and alarm. Med. Biol. Eng. Comput. 2007, 45, 877–885. [CrossRef] 114. Cho, G.; Jeong, K.; Paik, M.J.; Kwun, Y.; Sung, M. Performance evaluation of textile-based electrodes and motion sensors for smart clothing. IEEE Sens. J. 2011, 11, 3183–3193. [CrossRef] 115. Pola, T.; Vanhala, J. Textile electrodes in ECG measurement. In Proceedings of the 2007 3rd International Conference on Intelligent Sensors, Sensor Networks and Information, Melbourne, QLD, Australia, 3–6 December 2007; pp. 635–639. 116. Paul, G.; Torah, R.; Beeby, S.; Tudor, J. The development of screen-printed conductive networks on textiles for biopotential monitoring applications. Sens. Actuator A Phys. 2014, 206, 35–41. [CrossRef] 117. Qin, H.; Li, J.; He, B.; Sun, J.; Li, L.; Qian, L. Novel Wearable Electrodes Based on Conductive Chitosan Fabrics and Their Application in Smart Garments. Materials 2018, 11, 370. [CrossRef][PubMed]