Abstract
Artificial neural networks are one of the most efficient methods for pattern recognition and have a vast range of applications for aiding medical decision making. The proposed study applies feed-forward back-propagation neural networks as a classifier and compares the combination of nine learning algorithms and three activation functions to build a knowledge-based system with the best network architecture for predicting the severity of autism. The performances of the derived models were evaluated based on statistical criteria such as mean squared error (MSE), mean absolute percentage error (MAPE), root mean squared error (RMSE), regression (R value), training time and number of epochs. The study findings showed that the optimal performance was achieved by model MLP_LM_104 trained on Levenberg–Marquardt (LM) back-propagation algorithm having network topology of 40-10-4 with purelin and tansig activation functions in hidden and output layers. The regression coefficients for training, validation and test datasets were 0.996, 0.996 and 0.994, respectively. The MSE, RMSE and MAPE were \( 2.26 \times 10^{ - 4} \), \( 1.50 \times 10^{ - 2} \) and \( 1.13 \), respectively. Furthermore, BFGS quasi-Newton (BFG), conjugate gradient, gradient descent and resilient back-propagation (RP) algorithms did not perform well. Models trained with BFG algorithms required longer training time, whereas the performance of models trained on RP algorithm got worse as the numbers of hidden neurons were increased.
Similar content being viewed by others
References
Ahmed FE (2005) Artificial neural networks for diagnosis and survival prediction in colon cancer. Mol Cancer 4:29. doi:10.1186/1476-4598-4-29
Ayat S, Farahani HA, Aghamohamadi M, Alian M, Aghamohamadi S, Kazemi Z (2013) A comparison of artificial neural networks learning algorithms in predicting tendency for suicide. Neural Comput Appl 23:1381–1386. doi:10.1007/s00521-012-1086-z
Bai VT, Srivatsa SK (2006) Wireless tele care system for intensive care unit of hospitals using bluetooth and embedded technology. Inform Technol J 5:1106–1112. doi:10.3923/itj.2006.1106.1112
Bailey A, Le Couteur A, Gottesman I, Bolton P, Simonoff E, Yuzda E, Rutter M (1995) Autism as a strongly genetic disorder: evidence from a British twin study. Psychol Med 25:63–77. doi:10.1017/s0033291700028099
Baird G, Charman T, Cox A, Baron-Cohen S, Swettenham J, Wheelwright S, Drew A (2001) Current topic: screening and surveillance for autism and pervasive developmental disorders. Arch Dis Child 84:468–475. doi:10.1136/adc.84.6.468
Bartosch-Härlid A, Andersson B, Aho U, Nilsson J, Andersson R (2008) Artificial neural networks in pancreatic disease. Br J Surg 95:817–826. doi:10.1002/bjs.6239
Basheer IA, Hajmeer M (2000) Artificial neural networks: fundamentals, computing, design and application. J Microbiol Meth 43:3–31. doi:10.1016/S0167-7012(00)00201-3
Belmonte MK, Cook EH, Anderson GM, Rubenstein JLR, Greenough WT, Beckel-Mitchener A, Courchesne E, Boulanger LM, Powell SB, Levitt PR, Perry EK, Jiang YH, DeLorey TM, Tierney E (2004) Autism as a disorder of neural information processing: directions for research and targets for therapy. Mol Psychiatry 9:646–663. doi:10.1038/sj.mp.4001499
Bhatt A (2002) Reservoir properties from well logs using neural networks. Dissertation, Norwegian University of Science and Technology
Bishop CM (1995) Neural networks for pattern recognition. Clarendon Press, Oxford
Bouabaz M, Hamami M (2008) A cost estimation model for repair bridges based on artificial neural network. Am J Appl Sci 5:334–339. doi:10.3844/ajassp.2008.334.339
Bradley BP (2012) Finding biomarkers is getting easier. Ecotoxicology 21:631–636. doi:10.1007/s10646-011-0848-1
Broyden CG (1970) The convergence of a class of double-rank minimization algorithms. IMA J Appl Math 6:222–231. doi:10.1093/imamat/6.3.222
Bryson SE, Rogers SJ, Fombonne E (2003) Autism spectrum disorders: early detection, intervention, education, and psychopharmacological management. Can J Psychiatry 48:506–516
Caruana R, Lawrence S, Giles CL (2001) Overfitting in neural networks: backpropagation, conjugate gradient and early stopping. In: Leen TK, Dietterich TG, Tresp V (eds) Advances in neural information processing systems 13. MIT Press, Cambridge
Caudill M, Butler C (1992) Understanding neural networks, volume 1: basic networks. The MIT Press, Cambridge
Charman T, Taylor E, Drew A, Cockerill H, Brown JA, Baird G (2005) Outcome at 7 years of children diagnosed with autism at age 2: predictive validity of assessments conducted at 2 and 3 years of age and pattern of symptom change over time. J Child Psychol Psychiatry 46:500–513. doi:10.1111/j.1469-7610.2004.00377.x
Cheng B, Titterington DM (1994) Neural networks: a review from a statistical perspective. Stat Sci 9:2–30. doi:10.1214/ss/1177010638
Cuccaro ML, Wright HH, Rownd CV, Abramson RK, Waller J, Fender D (1996) Professional perceptions of children with developmental difficulties: the influence of race and socioeconomic status. J Autism Dev Disord 26:461–469. doi:10.1007/bf02172830
Daliakopoulos NI, Couliblay P, Tsanis HK (2005) Ground water level forcasting using artificial neural networks. J Hydrol 309:229–240
Dearlove J, Kearney D (1990) How good is general practice developmental screening? BMJ 300:1177–1180. doi:10.1136/bmj.300.6733.1177
Dee GL, Bakhary N, Rahman AA, Ahmad BH (2011) A comparison of artificial neural network learning algorithms for vibration-based damage detection. J Adv Mater Res 163–167:2756–2760. doi:10.4028/www.scientific.net/AMR.163-167.2756
Demuth H, Beale M, Hagan M (2008) Neural network toolbox™ 6. User’s guide
Dobos AE, Dworkin PH, Bernstein BA (1994) Pediatricians’ approaches to developmental problems: has the gap been narrowed? J Dev Behav Pediatr 15:34–38. doi:10.1097/00004703-199402000-00006
Dorofki M, Elshafie AH, Jaafar O, Karim OA, Mastura S (2012) Comparison of artificial neural network transfer functions abilities to simulate extreme runoff data. Int Proc Chem Biol Environ Eng 33:39–44
Elveren E, Yumuşak N (2011) Tuberculosis disease diagnosis using artificial neural network trained with genetic algorithm. J Med Syst 35:329–332. doi:10.1007/s10916-009-9369-3
Er O, Temurtas F, Tanrikulu AC (2010) Tuberculosis disease diagnosis using artificial neural networks. J Med Syst 34:299–302. doi:10.1007/s10916-008-9241-x
Erzin Y, Gul TO (2014) The use of neural networks for the prediction of the settlement of one-way footings on cohesionless soils based on standard penetration test. Neural Comput Appl 24:891–900. doi:10.1007/s00521-012-1302-x
Farahani HA, Kazemi Z, Aghamohamadi S, Ansari M, Aghamohamadi M (2011) Predicting tendency for suicide based on risk factors and ontological identity using artificial neural network. Psychol Res 1:227–238
Fletcher R, Reeves CM (1964) Function minimization by conjugate gradients. Comput J 7:149–154. doi:10.1093/comjnl/7.2.149
Fombonne E (2005) Epidemiology of autistic disorder and other pervasive developmental disorders. J Clin Psychiatry 10:3–8
Ghaffari A, Abdollahi H, Khoshayand MR, Bozchalooi IS, Dadgar A, Rafiee-Tehrani M (2006) Performance comparison of neural network training algorithms in modeling of bimodal drug delivery. Int J Pharm 327:126–138. doi:10.1016/j.ijpharm.2006.07.056
Gill PE, Murray W, Wright MH (1981) Practical optimization. Academic Press, New York
Glascoe FP (1997) Parents’ concerns about children’s development: prescreening technique or screening test? Pediatrics 99:522–528. doi:10.1542/peds.99.4.522
Gupta MK, Agarwal K, Prakash N, Singh DB, Misra K (2012) Prediction of miRNA in HIV-1 genome and its targets through artificial neural network: a bioinformatics approach. Netw Model Anal Health Inform Bioinform 1:141–151. doi:10.1007/s13721-012-0017-3
Hagan MT, Menhaj MB (1994) Training feedforward networks with the Marquardt algorithm. IEEE Trans Neural Netw 5:989–993. doi:10.1109/72.329697
Hager WW, Zhang H (2006) A survey of nonlinear conjugate gradient methods. Pacific J Optim 2:35–58
Heidgerken AD, Geffken G, Modi A, Frakey L (2005) A survey of autism knowledge in a health care setting. J Autism Dev Disord 35:323–330. doi:10.1007/s10803-005-3298-x
Howlin P (1998) Children with autism and asperger syndrome. A guide for practitioners and carers. John Wiley & Sons, Chichester
Ince R (2004) Prediction of fracture parameters of concrete by artificial neural networks. Eng Fract Mech 71:2143–2159. doi:10.1016/j.engfracmech.2003.12.004
Iranmanesh S, Mahdavi MA (2009) A differential adaptive learning rate method for back-propagation neural networks. World Acad Sci Eng Technol 50:285–288
Kermani BG, Schiffman SS, Nagle HT (2005) Performance of the Levenberg–Marquardt neural network training method in electronic nose applications. Sens Actuators B Chem 110:13–22. doi:10.1016/j.snb.2005.01.008
Lahmiri S (2011) A comparative study of backpropagation algorithms in financial prediction. Int J Comput Sci Eng Appl. doi:10.5121/ijcsea.2011.1402
Lawrence J (1994) Introduction to neural networks: design, theory and applications. California Scientific Software Press, California
Lin RH (2009) An intelligent model for liver disease diagnosis. Artif Intell Med 47:53–62. doi:10.1016/j.artmed.2009.05.005
Lord C, Risi S, DiLavore PS, Shulman C, Thurm A, Pickles A (2006) Autism from 2 to 9 years of age. Arch Gen Psychiatry 63:694–701. doi:10.1001/archpsyc.63.6.694
Lord C, Petkova E, Hus V, Gan W, Lu F, Martin DM, Ousley O, Guy L, Bernier R, Gerdts J, Algermissen M, Whitaker A, Sutcliffe JS, Warren Z, Klin A, Saulnier C, Hanson E, Hundley R, Piggot J, Fombonne E, Steiman M, Miles J, Kanne SM, Goin-Kochel RP, Peters SU, Cook EH, Guter S, Tjernagel J, Green-Snyder LA, Bishop S, Esler A, Gotham K, Luyster R, Miller F, Olson J, Richler J, Risi S (2012) A multisite study of the clinical diagnosis of different autism spectrum disorders. Arch Gen Psychiatry 69:306–313. doi:10.1001/archgenpsychiatry.2011.148
Magoulas GD, Vrahatis MN, Androulakis GS (1999) Improving the convergence of the backpropagation algorithm using learning rate adaptation methods. Neural Comput 11:1769–1796. doi:10.1162/089976699300016223
Mandell DS, Novak MM, Zubritsky CD (2005) Factors associated with age of diagnosis among children with autism spectrum disorders. Pediatrics 116:1480–1486. doi:10.1542/peds.2005-0185
Maria PRF (2002) An intelligent, interactive, web-based platform for effective clinical education through VR and distance learning modules. Inform Technol J 1:126–131. doi:10.3923/itj.2002.126.131
Mazurowski MA, Habas PA, Zurada JM, Lo JY, Baker JA, Tourassi GD (2008) Training neural network classifiers for medical decision making: the effects of imbalanced datasets on classification performance. Neural Netw 21:427–436. doi:10.1016/j.neunet.2007.12.031
Modai I, Saban NI, Stoler M, Valevski A, Saban N (1995) Sensitivity profile of 41 psychiatric parameters determined by neural network in relation to 8-week outcome. Comput Hum Behav 11:181–190. doi:10.1016/0747-5632(94)00022-a
Moller MF (1993) A scaled conjugate gradient algorithm for fast supervised learning. Neural Networks 6:525–533. doi:10.1016/S0893-6080(05)80056-5
Nair J, Nair S, Kashani J, Reid J, Mistry S, Vargas V (1999) Analysis of the symptoms of depression—a neural network approach. J Psychiatry Res 87:193–201. doi:10.1016/s0165-1781(99)00054-2
Nocedal J, Wright SJ (2000) In: Mikosch TV, Resnich SI, Robinson SM (eds) Limited-memory quasi-Newton methods: numerical optimization, 2nd edn. Springer, New York
Nordahl CW, Dierker D, Mostafavi I, Schumann CM, Rivera SM, Amaral DG, Van Essen DC (2007) Cortical folding abnormalities in autism revealed by surface-based morphometry. J Neurosci 27:11725–11735. doi:10.1523/jneurosci.0777-07.2007
Noterdaeme M, Amorosa H, Mildenberger K, Sitter S, Minow F (2001) Evaluation of attention problems in children with autism and children with a specific language disorder. Eur Child Adolesc Psychiatry 10:58–66. doi:10.1007/s007870170048
Ohta M, Nagai Y, Hara H, Sasaki M (1987) Parental perception of behavioral symptoms in Japanese autistic children. J Autism Dev Disord 17:549–563. doi:10.1007/bf01486970
Pace F, Savarino V (2007) The use of artificial neural network in gastroenterology: the experience of the first 10 years. Eur J Gastroenterol Hepatol 19:1043–1045. doi:10.1097/MEG.0b013e3282f198e5
Pinto-Martin JA, Young LM, Mandell DS, Poghosyan L, Giarelli E, Levy SE (2008) Screening strategies for autism spectrum disorders in pediatric primary care. J Dev Behav Pediatr 29:345–350. doi:10.1097/dbp.0b013e31818914cf
Powell MJD (1977) Restart procedures for the conjugate gradient method. Math Program 12:241–254. doi:10.1007/BF01593790
Raju KS, Kumar DN, Duckstein L (2006) Artificial neural networks and multicriterion analysis for sustainable irrigation planning. Comput Oper Res 33:1138–1153. doi:10.1016/j.cor.2004.09.010
Rapin I (1997) Autism. N Engl J Med 337:97–104. doi:10.1056/nejm199707103370206
Riedmiller (1993) Proceedings of the IEEE International Conference on Neural Networks (ICNN). San Francisco pp 586–591
Rumelhart DE, Hinton GE, Williams RJ (1986) Learning representations by back-propagating errors. Nature 323:533–536. doi:10.1038/323533a0
Saini LM, Soni MK (2002) Artificial neural network based peak load forecasting using Levenberg-Marquardt and quasi-Newton methods. IEEE Proc Gener Transm Dist 149:578–584. doi:10.1049/ip-gtd:20020462
Scales LE (1985) Introduction to non-linear optimization. Springer, Berlin
Setiono R, Hui LCK (1995) Use of a quasi-Newton method in a feedforward neural network construction algorithm. IEEE Trans Neural Netw 6:273–277. doi:10.1109/72.363426
Shah K (2001) What do medical students know about autism? Autism 5:127–133. doi:10.1177/1362361301005002003
Shankaracharya Odedra D, Samanta S, Vidyarthi AS (2010) Computational intelligence in early diabetes diagnosis: a review. Rev Diabet Stud 7:252–262. doi:10.1900/RDS.2010.7.252
Shevell MI, Majnemer A, Rosenbaum P, Abrahamowicz M (2001) Profile of referrals for early childhood developmental delay to ambulatory subspecialty clinics. J Child Neurol 16:645–650. doi:10.1177/088307380101600904
Sices L, Feudtner C, McLaughlin J, Drotar D, Williams M (2003) How do primary care physicians identify young children with developmental delays? A national survey. J Dev Behav Pediatr 24:409–417. doi:10.1542/peds.113.2.274
Siristatidis CS, Chrelias C, Pouliakis A, Katsimanis E, Kassanos D (2010) Artificial neural networks in gynaecological diseases: current and potential future applications. Med Sci Monit 16:231–236
Spelt L, Andersson B, Nilsson J, Andersson R (2012) Prognostic models for outcome following liver resection for colorectal cancer metastases: a systematic review. Eur J Surg Oncol 38:16–24. doi:10.1016/j.ejso.2011.10.013
Stevens R, Ikeda J, Casillas A, Palacio-Cayetano J, Clyman S (1999) Artificial neural network-based performance assessments. Comput Hum Behav 15(3–4):295–313. doi:10.1016/s0747-5632(99)00025-4
Talaee PH (2014) Multilayer perceptron with different training algorithms for streamflow forecasting. Neural Comput Appl 24:695–703. doi:10.1007/s00521-012-1287-5
Temurtas H, Yumusak N, Temurtas F (2009) A comparative study on diabetes disease diagnosis using neural networks. Expert Syst Appl 36:8610–8615. doi:10.1016/j.eswa.2008.10.032
Thakur A, Mishra V, Jain S (2011) Feed forward artificial neural network: tool for early detection of ovarian cancer. Sci Pharm 79:493–505. doi:10.3797/scipharm.1105-11
Uğuz H (2012) A biomedical system based on artificial neural network and principal component analysis for diagnosis of the heart valve diseases. J Med Syst 36:61–72. doi:10.1007/s10916-010-9446-7
Wassink TH, Brzustowicz LM, Bartlett CW, Szatmari P (2004) The search for autism disease genes. Ment Retard Dev Disabil Res Rev 10:272–283. doi:10.1002/mrdd.20041
Weigend A (1993) On overfitting and the effective number of hidden units. In: Mozer A, Smolensky P, Touretzky DS, Elman JL, Weigend AS (eds) Proceedings of the 1993 connectionist models summer school Hillsdale. Lawrence Erlbaum Associates, NJ, pp 335–342
Wiggins LD, Baio J, Rice C (2006) Examination of the time between first evaluation and first autism spectrum diagnosis in a population-based sample. J Dev Behav Pediatr 27:79–87. doi:10.1097/00004703-200604002-00005
Wilding P, Morgan M, Grygotis A, Shoffner M, Rosato E (1994) Application of backpropagation neural networks to diagnosis of breast and ovarian cancer. Cancer Lett 77:145–155. doi:10.1016/0304-3835(94)90097-3
Yeargin-Allsopp M, Rice C, Karapurkar T, Doernberg N, Boyle C, Murphy C (2003) Prevalence of autism in a US metropolitan area. J Am Med Assoc 289:49–55. doi:10.1001/jama.289.1.49
Acknowledgments
The authors wish to thank National Institute for the Mentally Handicapped (NIMH) for providing the autistic patients’ data and to the Dept. of Bioinformatics, Karunya University for their constant support and motivation.
Conflict of interest
The authors declare that they have no conflict of interest.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Chand, Y., Alam, M.A. & Tejaswini, Y.R.S.N. Performance comparison of artificial neural networks learning algorithms and activation functions in predicting severity of autism. Netw Model Anal Health Inform Bioinforma 4, 2 (2015). https://doi.org/10.1007/s13721-014-0073-y
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s13721-014-0073-y