Skip to main content

Advertisement

SpringerLink
Log in

Fast neural network learning algorithms for medical applications

  • Original Article
  • Published:
Neural Computing and Applications Aims and scope Submit manuscript

Abstract

Measuring the blood urea nitrogen concentration is crucial to evaluate dialysis dose (Kt/V) in patients with renal failure. Although frequent measurement is needed to avoid inadequate dialysis efficiency, artificial intelligence can repeatedly perform the forecasting tasks and may be a satisfactory substitute for laboratory tests. Artificial neural networks represent a promising alternative to classical statistical and mathematical methods to solve multidimensional nonlinear problems. It also represents a promising forecasting application in nephrology. In this study, multilayer perceptron (MLP) neural network with fast learning algorithms is used for the accurate prediction of the post-dialysis blood urea concentration. The capabilities of eight different learning algorithms are studied, and their performances are compared. These algorithms are Levenberg–Marquardt, resilient backpropagation, scaled conjugate gradient, conjugate gradient with Powell–Beale restarts, Polak–Ribiere conjugate gradient and Fletcher–Reeves conjugate gradient algorithms, BFGS quasi-Newton, and one-step secant. The results indicated that BFGS quasi-Newton and Levenberg–Marquardt algorithm produced the best results. Levenberg–Marquardt algorithm outperformed clearly all the other algorithms in the verification phase and was a very robust algorithm in terms of mean absolute error (MAE), root mean square error (RMSE), Pearson’s correlation coefficient (\( R_{p}^{2} \)) and concordance coefficient (R C ). The percentage of MAE and RMSE for Levenberg–Marquardt is 0.27 and 0.32 %, respectively, compared to 0.38 and 0.41 % for BFGS quasi-Newton and 0.44 and 0.48 % for resilient backpropagation. MLP-based systems can achieve satisfying results for predicting post-dialysis blood urea concentration and single-pool dialysis dose sp Kt/V without the need of a detailed description or formulation of the underlying process in contrast to most of the urea kinetic modeling techniques.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7

Similar content being viewed by others

References

  1. Acchiardo SR, Hatten KW, Ruvinsky MJ, Dyson B, Fuller J, Moore LW (1992) Inadequate dialysis increases gross mortality rate. ASAIO J 38(3):M282–M285

    Google Scholar 

  2. Ahrenholz P, Taborsky P, Bohling M et al (2011) Determination of dialysis dose: a clinical comparison of methods. Blood Purif 32(4):271–277

    Google Scholar 

  3. Akl AI, Sobh MA, Enab YM, Tattersall J (2001) Artificial intelligence: a new approach for prescription and monitoring of hemodialysis therapy. Am J Kidney Dis 38(6):1277–1283

    Google Scholar 

  4. Alloatti S, Molino A, Manes M, Bosticardo GM (1998) Urea rebound and effectively delivered dialysis dose. Nephrol Dial Transplant 13(Suppl 6):25–30

    Google Scholar 

  5. Azar AT, Wahba KM (2011) Artificial neural network for prediction of equilibrated dialysis dose without intradialytic sample. Saudi J Kidney Dis Transpl 22(4):705–711

    Google Scholar 

  6. Azar AT, Balas VE, Olariu T (2010) Artificial neural network for accurate prediction of post-dialysis urea rebound. doi:10.1109/SOFA.2010.5565606

  7. Azar AT (2008) Estimation of accurate and new method for hemodialysis dose calculation. Clin Med Insights Urol Issue 1:15–21

    Google Scholar 

  8. Barron AR (1993) Universal approximation bounds for superpositions of a sigmoid function. IEEE Trans Inf Theory 39(3):930–945

    MathSciNet  MATH  Google Scholar 

  9. Baroni M, Clementi S, Cruciani G, Costantino G, Rignanelli D, Oberrauch E (1992) Predictive ability of regression models: part II. Selection of the best predictive PLS model. J Chemom 6(6):347–356

    Google Scholar 

  10. Barth RH (1988) Direct calculation of Kt/V: a simplified approach to monitoring of haemodialysis. Nephron 50(3):191–195

    Google Scholar 

  11. Basile C, Vernaglione L, Lomonte C et al (2010) A new index of hemodialysis adequacy: clearance × dialysis time/bioelectrical resistance. J Nephrol 23(5):575–586

    Google Scholar 

  12. Basile C, Vernaglione L, Lomonte C et al (2010) Comparison of alternative methods for scaling dialysis dose. Nephrol Dial Transplant 25(4):1232–1239

    Google Scholar 

  13. Basile C, Casino F, Lopez T (1990) Percent reduction in blood urea concentration during dialysis estimates Kt/V in a simple and accurate way. Am J Kidney Dis 15(1):40–45

    Google Scholar 

  14. Battiti R, Masulli F (1990) BFGS optimization for faster and automated supervised learning, INCC 90 Paris, international neural network conference, pp 757–760

  15. Battiti R (1992) First and second order methods for learning: between steepest descent and Newton’s method. Neural Comput 4(2):141–166

    Google Scholar 

  16. Beale MH, Hagan MT, Demuth HB (2011) Neural network toolbox™ 7 user’s guide. The MathWorks, Inc., Natick

    Google Scholar 

  17. Beale EML (1972) A derivation of conjugate gradients. In: Lootsma FA (ed) Numerical methods for nonlinear optimization. Academic Press, London

    Google Scholar 

  18. Berry MJA, Linoff G (1997) Data mining techniques. Wiley, New York

    Google Scholar 

  19. Bland JM, Altman DG (1986) Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 8:307–310

    Google Scholar 

  20. Blum A (1992) Neural networks in C++. Wiley, New York

    Google Scholar 

  21. Blum A, Langley P (1997) Selection of relevant features and examples in machine learning. Artif Intell 97(1–2):245–271

    MathSciNet  MATH  Google Scholar 

  22. Boger Z, Guterman H (1997) Knowledge extraction from artificial neural network models, IEEE systems, man, and cybernetics conference, Orlando, FL

  23. Breitsameter G, Figueiredo AE, Kochhann DS (2012) Calculation of Kt/V in haemodialysis: a comparison between the formulas. J Bras Nefrol 34(1):22–26

    Google Scholar 

  24. Bridle JS (1989) Probabilistic interpretation of feedforward classification network outputs, with relationships to statistical pattern recognition. In: Fougelman-Soulie F (ed) Neurocomputing: algorithms, architectures and applications. Springer, Berlin, pp 227–236

    Google Scholar 

  25. Burgelman M, Vanholder R, Fostier H, Ringoir S (1997) Estimation of parameters in a two-pool urea kinetic model for hemodialysis. Med Eng Phys 19(1):69–76

    Google Scholar 

  26. Calzavara P, Vianello A, da Porto A, Gatti PL, Bertolone G, Caenaro G, Dalla Rosa C (1988) Comparison between three mathematical models Of Kt/V. Int J Artif Organs 11(2):107–110

    Google Scholar 

  27. Camps-Valls G, Porta-Oltra B, Soria-Olivas E, Martin-Guerrero JD, Serrano-Lopez AJ, Perez-Ruixo JJ, Jimenez-Torres NV (2003) Prediction of cyclosporine dosage in patients after kidney transplantation using neural networks. IEEE Trans Biomed Eng 50(4):442–448

    Google Scholar 

  28. Collins AJ, Ma JZ, Umen A, Keshaviah P (1994) Urea index and other predictors of hemodialysis patient survival. Am J Kidney Dis 23(2):272–282

    Google Scholar 

  29. Cross SS, Harrison RF, Kennedy RL (1995) Introduction to neural networks. Lancet 346(8982):1075–1079

    Google Scholar 

  30. Daugirdas JT, Leypoldt JK, Akonur A et al. (2012) Improved equation for estimating single-pool Kt/V at higher dialysis frequencies. Nephrol Dial Transplant (in print). doi:10.1093/ndt/gfs115

  31. Daugirdas JT (2011) Is there a minimal amount of time a patient should be dialyzed regardless of measured KT/V? Semin Dial 24(4):423–425

    Google Scholar 

  32. Daugirdas JT, Greene T, Depner TA, Levin NW, Chertow GM (2011) Modeled urea distribution volume and mortality in the HEMO study. Clin J Am Soc Nephrol 6(5):1129–1138

    Google Scholar 

  33. Daugirdas JT, Greene T, Chertow GM, Depner TA (2010) Can rescaling dose of dialysis to body surface area in the HEMO study explain the different responses to dose in women versus men? Clin J Am Soc Nephrol 5(9):1628–1636

    Google Scholar 

  34. Daugirdas JT, Depner TA, Greene T et al (2010) Standard Kt/Vurea: a method of calculation that includes effects of fluid removal and residual kidney clearance. Kidney Int 77(7):637–644

    Google Scholar 

  35. Daugirdas JT (1989) The post: pre dialysis plasma urea nitrogen ratio to estimate Kt/V and nPCR: validation. Int J Artif Organs 12(7):420–427

    Google Scholar 

  36. Daugirdas JT (1993) Second generation logarithmic estimates of single-pool variable volume Kt/V: an analysis of error. J Am Soc Nephrol 4(5):1205–1213

    Google Scholar 

  37. Daugirdas JT, Schneditz D, Leehey DJ (1996) Effect of access recirculation on the modeled urea distribution volume. Am J Kidney Dis 27(4):512–518

    Google Scholar 

  38. Daugirdas JT, Greene T, Depner TA, Leypoldt J, Gotch F, Schulman G, Star R (2004) Factors that affect postdialysis rebound in serum urea concentration, including the rate of dialysis: results from the HEMO Study. J Am Soc Nephrol 15(1):194–203

    Google Scholar 

  39. Dennis JE, Schnabel RB (1983) Numerical methods for unconstrained optimization and nonlinear equations. Prentice-Hall, Englewood Cliffs

    MATH  Google Scholar 

  40. Depner TA (1990) Prescribing hemodialysis: a guide to urea modeling, 2nd edn. Springer, Berlin

    Google Scholar 

  41. Depner TA (1994) Assessing adequacy of hemodialysis urea modeling. Kidney Int 45(5):1522–1535

    Google Scholar 

  42. Depner TA (1996) Quantifying hemodialysis. Am J Nephrol 16(1):17–28

    Google Scholar 

  43. Depner TA (1999) History of dialysis quantitation. Semin Dial 12(1):S14–S19

    Google Scholar 

  44. Depner TA (2005) Hemodialysis adequacy: basic essentials and practical points for the nephrologist in training. Hemodial Int 9(3):241–254

    Google Scholar 

  45. DeVore RA, Howard R, Micchelli CA (1989) Optimal nonlinear approximation. Manuscripta Math 63(4):469–478

    MathSciNet  Google Scholar 

  46. Fernandez de Canete J, Del Saz Huang P (2010) First-principles modeling of fluid and solute exchange in the human during normal and hemodialysis conditions. Comput Biol Med 40(9):740–750

    Google Scholar 

  47. Fernandez EA, Valtuille R, Willshaw P, Perazzo CA (2001) Using artificial intelligence to predict the equilibrated post-dialysis blood urea concentration. Blood Purif 19(3):271–285

    Google Scholar 

  48. Fernandez EA, Valtuille R, Willshaw P, Perazzo CA (2003) Dialysate-side urea kinetics. Neural network predicts dialysis dose during dialysis. Med Biol Eng Comput 41(4):392–396

    Google Scholar 

  49. Fernandez EA, Valtuille R, Presedo J, Willshaw P (2005) Comparison of different methods for hemodialysis evaluation by means of ROC curves: from artificial intelligence to current methods. Clin Nephrol 64(3):205–213

    Google Scholar 

  50. Fernandez EA, Valtuille R, Rodriguez Presedo J, Willshaw P (2005) Comparison of standard and artificial neural network estimators of hemodialysis adequacy. Artif Organs 29(2):159–165

    Google Scholar 

  51. Fletcher R, Reeves CM (1964) Function minimization by conjugate gradients. Comput J 7(2):149–154

    MathSciNet  MATH  Google Scholar 

  52. Fletcher R (2000) Practical methods of optimization, 2nd edn. Wiley, New York

    Google Scholar 

  53. Furness PN, Kazi J, Levesley J, Taub N, Nicholson M (1999) A neural network approach to the diagnosis of early acute allograft rejection. Transplant Proc 31(8):3151

    Google Scholar 

  54. Furness PN, Levesley J, Luo Z, Taub N, Kazi JI, Bates WD, Nicholson ML (1999) A neural network approach to the biopsy diagnosis of early acute renal transplant rejection. Histopathology 35(5):461–467

    Google Scholar 

  55. Gabutti L, Ferrari N, Mombelli G, Marone C (2004) Does cystatin C improve the precision of Cockcroft and Gault’s creatinine clearance estimation? J Nephrol 17(5):673–678

    Google Scholar 

  56. Gabutti L, Burnier M, Mombelli G, Malé F, Pellegrini L, Marone C (2004) Usefulness of artificial neural networks to predict follow-up dietary protein intake in hemodialysis patients. Kidney Int 66(1):399–407

    Google Scholar 

  57. Gabutti L, Vadilonga D, Mombelli G, Burnier M, Marone C (2004) Artificial neural networks improve the prediction of Kt/V, follow-up dietary protein intake and hypotension risk in haemodialysis patients. Nephrol Dial Transplant 19(5):1204–1211

    Google Scholar 

  58. Gabutti L, Machacek M, Marone C, Ferrari P (2005) Predicting intradialytic hypotension from experience, statistical models and artificial neural networks. J Nephrol 18(4):409–416

    Google Scholar 

  59. Garred LJ, Canaud B, Bosc JY, Tetta C (1997) Urea rebound and delivered Kt/V determination with a continuous urea sensor. Nephrol Dial Transplant 12(3):535–542

    Google Scholar 

  60. Geddes CC, Fox JG, Allison ME, Boulton-Jones JM, Simpson K (1998) An artificial neural network can select patients at high risk of developing progressive IgA nephropathy more accurately than experienced nephrologists. Nephrol Dial Transplant 13(1):67–71

    Google Scholar 

  61. Gill PE, Murray W, Wright MH (1982) Practical optimization. Emerald Group Publishing Limited, UK

    Google Scholar 

  62. Goldfarb-Rumyantzev A, Schwenk MH, Liu S, Charytan C, Spinowitz BS (2003) Prediction of single-pool Kt/v based on clinical and hemodialysis variables using multilinear regression, tree-based modeling, and artificial neural networks. Artif Organs 27(6):544–554

    Google Scholar 

  63. Gotch FA (1990) Kinetic modeling in hemodialysis. In: Nissenson AR, Fine RN, Gentile DE (eds) Clinical dialysis, 2nd edn. Appleton & Lange, Norwalk, pp 118–146

    Google Scholar 

  64. Gotch FA (2001) Evolution of the single-pool urea kinetic model. Semin Dial 14(4):252–256

    Google Scholar 

  65. Gotch FA, Sargent JA (1985) A mechanistic analysis of the national cooperative dialysis study. Kidney Int 28(3):526–538

    Google Scholar 

  66. Grzegorzewska AE, Banachowicz W (2008) Evaluation of hemodialysis adequacy using online Kt/V and single-pool variable-volume urea Kt/V. Int Urol Nephrol 40(3):771–778

    Google Scholar 

  67. Guh JY, Yang CY, Yang JM, Chen LM, Lai YH (1998) Prediction of equilibrated postdialysis BUN by an artificial neural network in high-efficiency hemodialysis. Am J Kidney Dis 31(4):638–646

    Google Scholar 

  68. Gurney K, Wright MJ (1997) An introduction to neural networks. UCL Press (Taylor & Francis group), London

    Google Scholar 

  69. Guyon I, Elisseeff A (2003) An introduction to variable and feature selection. J Mach Leran Res 3:1157–1182

    MATH  Google Scholar 

  70. Hakim RM, Breyer J, Ismail N, Schulman G (1994) Effects of dose of dialysis on morbidity and mortality. Am J Kidney Dis 23(5):661–669

    Google Scholar 

  71. Hagan MT, Demuth HB, Beale MH (2002) Neural network design. PWS Publishing, Boston

    Google Scholar 

  72. Hagan MT, Menhaj M (1994) Training feed-forward networks with the Marquardt algorithm. IEEE Trans Neural Netw 5(6):989–993

    Google Scholar 

  73. Haykin S (1999) Neural networks, 2nd edn. Prentice Hall, New Jersey

    MATH  Google Scholar 

  74. Held PJ, Port FK, Wolfe RA, Stannard DC, Carroll CE, Daugirdas JT, Bloembergen WE, Greer JW, Hakim RM (1996) The dose of haemodialysis and patient mortality. Kidney Int 50(2):550–556

    Google Scholar 

  75. Henning MR (2007) Affecting Kt/V: an analysis of staff interventions. Dial Transplant 36(11):584–601

    Google Scholar 

  76. Hestenes M (1980) Conjugate direction methods in optimization. Springer, New York

    MATH  Google Scholar 

  77. Heston TF, Norman DJ, Barry JM, Bennett WM, Wilson RA (1997) Cardiac risk stratification in renal transplantation using a form of artificial intelligence. Am J Cardiol 79(4):415–417

    Google Scholar 

  78. Ijely GK, Raja RM (1991) Simplified calculation of PCR and Kt/V. Abstract 24th Annual JASN Meeting, p 329

  79. Jean G, Chazot C, Charra B, Terrat JC, Vanel T, Calemard E, Laurent G (1998) Is post-dialysis urea rebound significant with long slow hemodialysis? Blood Purif 16(4):187–196

    Google Scholar 

  80. Jean G, Charra B, Chazot C, Laurent G (1999) Quest for post-dialysis urea rebound-equilibrated Kt/V with only intradialytic urea samples. Kidney Int 56(3):1149–1153

    Google Scholar 

  81. Jindal KK, Manuel A, Goldstein MB (1987) Percent reduction in blood urea concentration during hemodialysis (PRU). A simple and accurate method to estimate Kt/V urea. ASAIO Trans 33(3):286–288

    Google Scholar 

  82. Johansson EM, Dowla FU, Goodman DM (1990) Backpropagation learning for multi-layer feed-forward neural networks using the conjugate gradient method. Int J Neural Syst (IJNS) 2(4):291–301

    Google Scholar 

  83. Jordan MI (1995) Why the logistic function? A tutorial discussion on probabilities and neural networks. MIT computational cognitive science report 9503. http://www.cs.berkeley.edu/*jordan

  84. Kazi JI, Furness PN, Nicholson M, Ahmed E, Akhter F, Naqvi A, Rizvi A (1999) Interinstitutional variation in the performance of Baysian belief network for the diagnosis of acute renal graft rejection. Transplant Proc 31(8):3152

    Google Scholar 

  85. Kerr PG, Argiles A, Canaud B, Flavier JL, Mion CM (1993) Accuracy of Kt/V estimations in high-flux hemodiafiltration using percent reduction of urea: incorporation of urea rebound. Nephrol Dial Transplant 8(2):140–153

    Google Scholar 

  86. Keshaviah PR, Hanson GI, Berkseth RO, Collins AJ (1988) A simplified approach to monitoring in vivo therapy prescription. Trans Am Soc Artif Organs 34(3):620–622

    Google Scholar 

  87. Lai YH, Guh JY, Chen HC, Tsai JH (1995) Effects of different sampling methods for measurement of post dialysis blood urea nitrogen on urea kinetic modeling derived parameters in patients undergoing long-term hemodialysis. ASAIO J 41(2):211–215

    Google Scholar 

  88. Leblanc M, Charbonneau R, Lalumiere G, Cartier P, Deziel C (1996) Postdialysis urea rebound: determinants and influence on dialysis delivery in chronic hemodialysis patients. Am J Kidney Dis 27(2):253–261

    Google Scholar 

  89. Lawrence S, Giles CL, Tsoi AC (1997) Lessons in neural network training: overfitting may be harder than expected. In: Proceedings of the fourteenth national conference on artificial intelligence, AAAI-97. AAAI Press, Menlo Park, pp 540–545

  90. Lowrie EG, Laird NM, Parker TF et al (1981) Effect of the hemodialysis prescription on patient morbidity: report from the national cooperative dialysis study. N Engl J Med 305(20):1176–1181

    Google Scholar 

  91. Lowrie EG, Teehan BP (1983) Principles of prescribing dialysis therapy: implementing recommendations from the national co-operative dialysis study. Kidney Int 23(Suppl 13):S113–S122

    Google Scholar 

  92. Lowrie EG, Zhu X, Lew NL (1998) Primary associates or mortality among dialysis patients: trends and reassessment of Kt/V and urea reduction ratio as outcome-based measures of dialysis dose. Am J Kidney Dis 32(Suppl 4):S16–S31

    Google Scholar 

  93. Malmgren H, Borga M, Niklasson L (2000) Artificial neural networks in medicine and biology, perspectives in neural computing. Springer, Goteborg

    Google Scholar 

  94. Marquardt DW (1963) An algorithm for least-squares estimation of nonlinear parameters. SIAM J Appl Math 11(2):431–441

    MathSciNet  MATH  Google Scholar 

  95. MedCalc (2012) Medical calculator statistical software. http://www.medcalc.be. Accessed on Jan 2012

  96. Moller MF (1993) A scaled conjugate gradient algorithm for fast supervised learning. Neural Netw 6:525–533

    Google Scholar 

  97. Montano JJ, Palmer A (2003) Numeric sensitivity analysis applied to feedforward neural networks. Neural Comput Appl 2(2):119–125

    Google Scholar 

  98. Mitchell T (1997) Machine learning. The McGraw-Hill Companies, Inc., New York

    MATH  Google Scholar 

  99. NCSS (2012) Statistical and power analysis software. http://www.ncss.com. Accessed on Jan 2012

  100. NKF-K/DOQI (1997) Clinical practice guidelines for hemodialysis adequacy. Am J Kidney Dis 30(Suppl 2):S15–S57

    Google Scholar 

  101. NKF-K/DOQI (2001) Clinical practice guidelines for hemodialysis adequacy: update. Am J Kidney Dis 37(Suppl 1):S7–S64

    Google Scholar 

  102. NKF-K/DOQI (2006) Clinical practice guidelines for hemodialysis adequacy, update, 2006. Hemodialysis adequacy 2006 work group. Am J Kidney Dis 48(Suppl 1):S2–S90

    Google Scholar 

  103. Picard RR, Cook RD (1984) Cross-validation of regression models. J Am Stat Assoc 79(387):575–583

    MathSciNet  MATH  Google Scholar 

  104. Penny W, Frost D (1996) Neural networks in clinical medicine. Med Decis Mak 16(4):386–398

    Google Scholar 

  105. Powell M (1977) Restart procedures for the conjugate gradient method. Math Program 12(1):241–254

    MATH  Google Scholar 

  106. Prado M, Roa L, Palma A, Milán JA (2004) A novel mathematical method based on urea kinetic modeling for computing the dialysis dose. Comput Methods Programs Biomed 74(2):109–128

    Google Scholar 

  107. Reggia JA (1993) Neural computation in medicine. Artif Intell Med 5(2):143–157

    Google Scholar 

  108. Renal Physicians Association (1993) Clinical practice guideline: adequacy of hemodialyses. Renal Physicians Association, Washington

    Google Scholar 

  109. Riedmiller M, Braun H (1993) A direct adaptive method for faster backpropagation learning: the RPROP algorithm. In: Proceedings of the IEEE international conference on neural networks, vol 1, pp 586–591, San Francisco, CA. doi:10.1109/ICNN.1993.298623

  110. Robert CP, Casella G (2004) Monte Carlo statistical methods, 2nd edn. Springer, New York. ISBN 0-387-21239-6

    MATH  Google Scholar 

  111. Rumelhart DE, McClelland JL (1986) Parallel distributed processing: exploration in the microstructure of cognition. MIT Press, Cambridge

    Google Scholar 

  112. Sargent JA, Gotch FA (1980) Mathematic modeling of dialysis therapy. Kidney Int Suppl 10:S2–S10

    Google Scholar 

  113. Sargent J (1983) Control of dialysis by a single-pool urea model: the national cooperative dialysis study. Kidney Int 23(suppl 13):S19–S26

    Google Scholar 

  114. Sarle WS (1995) Stopped training and other remedies for overfitting. In: Proceedings of the 27th symposium on the interface of computing science and statistics, pp 352–360

  115. Sarle WS (1997) Neural network FAQ. Periodic posting to the Usenet newsgroup comp.ai.neural-nets. ftp://ftp.sas.com/pub/neural/FAQ.html

  116. Schalkoff RJ (1997) Artificial neural networks. McGraw-Hill International Editions

  117. Schneditz D, Daugirdas JT (2001) Compartment effects in hemodialysis. Semin Dial 14(4):271–277

    Google Scholar 

  118. Schneditz D, VanStone J, Daugirdas JT (1993) A regional blood circulation alternative to in-series two compartment urea kinetic modeling. ASAIO J 39(3):M573–M577

    Google Scholar 

  119. Schneditz D, Fariyike B, Osheroff R, Levin NW (1995) Is intercompartmental urea clearance during hemodialysis a perfusion term? A comparison of two pool urea kinetic models. J Am Soc Nephrol 6(5):1360–1370

    Google Scholar 

  120. Sheppard D, McPhee D, Darke C, Shrethra B, Moore R, Jurewitz A, Gray A (1999) Predicting cytomegalovirus disease after renal transplantation: an artificial neural network approach. Int J Med Inform 54(1):55–76

    Google Scholar 

  121. Sherrod PH (2012) DTREG predictive modeling software. www.dtreg.com. Accessed on Jan 2012

  122. Shoskes DA, Ty R, Barba L, Sender M (1998) Prediction of early graft function in renal transplantation using a computer neural network. Transplant Proc 30(4):1316–1317

    Google Scholar 

  123. Simic-Ogrizovic S, Furuncic D, Lezaic V, Radivojevic D, Blagojevic R, Djukanovic L (1999) Using ANN in selection of the most important variables in prediction of chronic renal allograft rejection progression. Transplant Proc 31(1–2):368

    Google Scholar 

  124. Smith BP, Ward RA, Brier ME (1998) Prediction of anticoagulation during hemodialysis by population kinetics and an artificial neural network. Artif Organs 22(9):731–739

    Google Scholar 

  125. Smye SW, Hydon PE, Will E (1993) An analysis of the single-pool urea kinetic model and estimation of errors. Phys Med Biol 38(1):115–122

    Google Scholar 

  126. Spiegel DM, Baker PL, Babcock S, Cantiguglia R, Klein M (1995) Hemodialysis urea rebound: the effect of increasing dialysis efficiency. Am J Kidney Dis 25(1):26–29

    Google Scholar 

  127. Swingler K (1996) Applying neural networks: a practical guide. Academic Press, London

    Google Scholar 

  128. Tangri N, Ansell D, Naimark D (2008) Predicting technique survival in peritoneal dialysis patients: comparing artificial neural networks and logistic regression. Nephrol Dial Transplant 23(9):2972–2981

    Google Scholar 

  129. Tetko IV, Livingstone DJ, Luik AI (1995) Neural network studies 1, comparison of overfitting and overtraining. J Chem Info Comp Sci 35(5):826–833

    Google Scholar 

  130. Trujillano J, March J, Sorribas A (2004) Methodological approach to the use of artificial neural networks for predicting results in medicine. Med Clin (Barc) 122(Suppl 1):59–67

    Google Scholar 

  131. Tsoukalas LH, Uhrig RE (1997) Fuzzy and neural approaches in engineering. Wiley, NY

    Google Scholar 

  132. Tu JV (1996) Advantages and disadvantages of using artificial neural networks versus logistic regression for predicting medical outcomes. J Clin Epidemiol 49(11):1225–1231

    Google Scholar 

  133. Vapnik VN (1998) Statistical learning theory. Wiley-Interscience, USA

    MATH  Google Scholar 

  134. Veng-Pedersen P, Modi NB (1992) Neural networks in pharmacodynamic modeling. Is current modelling practice of complex kinetic systems at a dead end? J Pharmacokinet Biopharm 20(4):397–412

    Google Scholar 

  135. Veng-Pedersen P, Modi NB (1993) Application of neural networks to pharmacodynamics. J Pharm Sci 82(9):918–926

    Google Scholar 

  136. Weigend A (1994) On overfitting and the effective number of hidden units. In: Proceedings of the 1993 connectionist models summer school, pp 335–342

  137. Wolf AV, Remp DG, Kiley JE, Currie GD (1951) Artificial kidney function; kinetics of hemodialysis. J Clin Invest 30(10):1062–1070

    Google Scholar 

  138. Xu QS, Liang YZ (2001) Monte Carlo cross validation. Chemom Intell Lab Syst 56(1):1–11

    MathSciNet  Google Scholar 

  139. Yashiro M, Watanabe H, Muso E (2004) Simulation of post-dialysis urea rebound using regional flow model. Clin Exp Nephrol 8(2):139–145

    Google Scholar 

  140. Zweiri YH, Whidborne JF, Sceviratne LD (2002) A three-term backpropagation algorithm. Neurocomputing 50:305–318

    Google Scholar 

Download references

Acknowledgments

I would like to highly appreciate and gratefully acknowledge Phillip H. Sherrod, software developer and consultant on predictive modeling, for his support and consultation during modeling process. The author thanks all medical staff at the nephrology Department in Ahmad Maher Teaching Hospital, Cairo, Egypt, for their invaluable support during the course of this study.

Author information

Authors and Affiliations

  1. Computer and Software Engineering Department, Faculty of Engineering, Misr University for Science and Technology (MUST), 6th of October City, Egypt

    Ahmad Taher Azar

Authors
  1. Ahmad Taher Azar
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Ahmad Taher Azar.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Azar, A.T. Fast neural network learning algorithms for medical applications. Neural Comput & Applic 23, 1019–1034 (2013). https://doi.org/10.1007/s00521-012-1026-y

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00521-012-1026-y

Keywords

Search

Navigation