Mathematical Modelling of Urethral and Similar Flows
Total Page：16
File Type：pdf, Size：1020Kb
Mathematical modelling of urethral and similar flows Stephen E. Glavin Department of Mathematics UCL A thesis submitted for the degree of Doctor of Philosophy Supervisors Professor Frank Smith and Professor Guo-Xiong Wu October 2011 I, Stephen Glavin, confirm that the work presented in this thesis is my own. Where information has been derived from other sources, I confirm that this has been indicated in the thesis. Abstract Flows in flexible tubes and vessels have been studied extensively in the past with particular application to the cardiovascular and respiratory systems. How- ever there have been few treatments of the lower urinary tract, which consists of the bladder and urethra. This thesis concentrates specifically on the urethra with the aim of giving insight into the evolving flow characteristics within the vessel and mechanical responses of the vessel which give rise to fluid structure interactions. Urethral modelling is an important area of research given the social and economic costs involved in lower urinary tract dysfunction. In the modelling, examination is given to slow and fast opening vessels where certain exact analytical solutions are found along with numerical results. Following this, fast and slow responses of the walls of the vessels are considered, where the response is defined as the relative change in cross-sectional area for relatively varying transmural pressure. These features are important for pathologies that alter the characteristics of the vessel wall such as bladder outlet obstruction. A change in the distensibility along the vessel resulting from pathologies or normal transition through the various sections of the urethra is studied both in terms of developing jump conditions based on a localised Euler region and also over a comparatively short length scale giving rise to the Burgers equa- tion; small amplitude instabilities are studied through the derivation of the KdV equation. Following on from these mostly two-dimensional treatments, three-dimensional systems are then studied. Consideration is given to the sec- ondary flow effects driven by the tortuosity of a vessel in three dimensions. We study cases of three-dimensional constriction, with main interest in the effects of benign prostate hyperplasia or urethral stricture on the flow, where pressure drops are demonstrated. Finally an appendix deals with the effects concerned with a wide population, focusing on an allied problem of consumer choice. Acknowledgements My gratitude goes to Professor Frank Smith for all his support, advice, guid- ance and encouragement. This work would not have been possible without his enthusiasm and expertise. I would also like to thank Professor Guo-Xiong Wu for his advice and discussions on modelling, to Professor Alan Cottenden for all his useful discussions on the biology and mechanics of the lower uri- nary tract, to Professor Jean-Marc Vanden Broeck for his expertise and advice on free-surface problems and to Dr Abhijit Sengupta for his supervision and enthusiasm during my internship. I am grateful to the Department of Mathematics and Department of Mechan- ical Engineering for their funding throughout this PhD and also to Unilever UK and the Knowledge Transfer Network in industrial mathematics (sponsered by EPSRC) for their funding during my six month internship at Unilever. Finally I would like to give my thanks to my wife, family and friends for their enouragement and support and also to the postgraduates of the Mathematics department at UCL, in particular Dr Alex White, Dr Hannah Fry and Dr Marios Tziannaros. Contents 1 Introduction 20 1.1 Relevantanatomyandbiomechanics. 21 1.2 Thestructureofthemaleurethra . 27 1.2.1 Preprostaticpart ...................... 27 1.2.2 Prostaticpart........................ 27 1.2.3 Membraneouspart ..................... 28 1.2.4 Spongiosepart ....................... 28 1.3 Thestructureofthefemaleurethra . 29 1.4 Aimofthethesis .......................... 30 1.5 Distensibility . 31 1.6 Typicalvalues............................ 33 1.7 Review................................ 34 1.8 Structureofthepresentthesis . 35 I The Physical Model 37 2 Derivation of the long-tube equations 38 2.1 Dimensional equations. 38 2.1.1 Massconservation. .. 41 2.1.2 Momentumbalance. .................... 41 2.1.3 Pressure-area relationship (wall law). 43 3 CONTENTS 4 2.1.4 Governing equations for the (u,A)system. 44 2.1.5 Characteristic system for constant β............ 45 2.2 Non-dimensionalisation . 46 2.3 Early times; the opening of the vessel. 47 2.3.1 Similarity solutions for abrupt openings. 48 2.3.2 Similarity solutions for gradual openings. 50 2.4 Short-scale alterations in tube properties and discontinuities in thelong-scale ............................ 55 2.4.1 ForwardmarchingRKscheme . 59 2.5 Summaryofchapter ........................ 63 3 Fast and slow responses 64 3.1 BehaviourofthePDEsystemnearthefront. 65 3.2 Frontbehaviourforgradualopenings. 68 3.3 Numerically marching backwards in the U, B System . 70 3.3.1 Rapid entrance openings (small k) . 71 3.3.2 Fast-response vessels (large n)............... 80 3.3.3 Slow-response vessels (small n) .............. 81 3.4 Large-n properties in any two-dimensional vessel. 81 3.4.1 Opening the gap with a general power law in time. 87 3.5 Slow-response vessels (small values of n) ............. 88 3.5.1 Two-dimensional case . 88 3.6 Summaryofchapter ........................ 91 4 Numerical studies, introduction of jumps and tortuosity 92 4.1 Finite-difference scheme . 92 4.1.1 Fronttracking........................ 94 4.1.2 Numerical results . 95 4.2 Long scale approximation with short scale change in distensibility.115 CONTENTS 5 4.2.1 Numerical solutions . 118 4.2.2 Numericalmethod .....................119 4.2.3 Case 1: Starting from an undisturbed cross-section. 120 4.2.4 Case 2: Perturbation about the theoretical steady state. 123 4.2.5 Examining disturbances such as those seen in Case 2. 123 4.2.6 Finite time singularities. 128 4.3 Summaryofchapter ........................131 5 Three-dimensional modelling 132 5.1 Three-dimensional effects over a relatively long length scale. 132 5.1.1 Three-dimensional perturbation about a two-dimensional solution............................134 5.1.2 Adding tortuosity. 136 5.2 Large-n properties in three-dimensional vessels. 140 5.2.1 Three-dimensional effects over a relatively long length scale..............................140 5.2.2 Three-dimensional thin-region flow. 146 5.3 Summaryofchapter ........................162 6 Three-dimensional constriction 163 6.1 Governingequations . .. .. .163 6.2 Analysis close to the point of minimum gap . 165 6.2.1 Coreflow ..........................165 6.2.2 Saddle point of the point of greatest constriction . 166 6.3 Numerical solutions . 168 6.3.1 Symmetriccases ......................169 6.3.2 Non-symmetriccases . .175 6.3.3 Gridtests ..........................178 6.4 Summaryofchapter ........................178 CONTENTS 6 II Heterogeneous population modelling 180 7 Describing the context 181 III Conclusion 183 8 Final Discussion 184 8.1 FutureWork.............................186 Appendices 188 A KdV-like behaviour 188 A.1 Introduction.............................188 A.2 KdVapproximation. .. .. .191 B Volatility in the consumer packaged goods industry - a simu- lation based study. 199 B.1 Introduction.............................199 B.1.1 Background . 202 B.2 TheMarketModel .........................205 B.3 Data.................................207 B.4 Validation . 209 B.4.1 Initialization . 210 B.4.2 Calibration . 211 B.4.3 Testing . 215 B.4.4 Thebenchmark .......................216 B.4.5 Notes on validation . 217 B.5 Simulation Setup . 219 B.6 Results................................220 B.6.1 MarketLevel ........................221 CONTENTS 7 B.6.2 Household Level . 224 B.7 Conclusion..............................230 B.8 Figures................................231 List of Figures 1.1 Sketch of the lower urinary tract for females viewed from the front. Labels are D, detrusor smooth muscle; T, trigone; SM, urethral smooth muscle; DS, distal intrinsic urethral sphincter; PS, periurethral sphincter; BN, bladder neck; O, ureteral ori- fices; C, connective tissue. This diagram is adapted from that foundin[40]. ............................ 24 1.2 Sketch of the lower urinary tract for males viewed from the front and left side respectively. Labels are D, detrusor smooth muscle; T, trigone; SM, smooth muscle; DS, distal intrinsic ure- thral sphincter; PS, periurethral sphincter; BN, bladder neck; P, prostate gland; MU, membraneous urethra; PU, penile urethra; EM, external meatus; E, ejaculatory duct; O, ureteral orifices. This diagram is adapted from that found in [40]. 25 1.3 Cross-sections of a human female urethra taken from [64] illus- trating the changes in shape of the urethra from the external meatus(a)tothebladderneck(f). 26 1.4 Sketch of a tube law (pressure-area relation) for a thin walled elastic tube with approximate (simplified) wall shapes. [40],[62],[71]. 32 8 LIST OF FIGURES 9 2.1 The “one-dimensional vessel”. Above, the original slowly vary- ing vessel (not to scale), and below, the one-dimensional orien- tation. Figure adapted from [71]. 39 2.2 Similarity solution showing gradual opening of an initially closed 1 vessel for n = 2 and β = 1 showing velocity (dashed line) and cross-sectional area (solid line) for t = 0, 0.1, 0.2,..., 1. Values at the vessel opening are a(0, t) = 1 and u(0, t) = 1 and are √2 foundfromequations(2.55)and(2.56). 51 2.3 Similarity solution showing gradual opening of an initially closed vessel for n = 2 and β = 1 showing velocity (dashed line) and cross-sectional area (solid line) for t = 0, 0.1, 0.2,..., 1. Values at the vessel opening are a(0, t) = 1 and u(0, t) = √2 and are foundfromequations(2.55)and(2.56). 52 2.4 Similarity solution showing gradual opening of an initially closed vessel for n = 100 and β = 1 showing velocity (dashed line) and cross-sectional area (solid line) for t = 0, 0.1, 0.2,..., 1. Values at the vessel opening are a(0, t) = 1 and u(0, t) = 10 and are foundfromequations(2.55)and(2.56). 53 2.5 Plot of a against η for the special case n = 2 with u = a for values of k = 0.01, 0.1, 1, 2, 10, 100 .