Diagnosis and prediction of obesity and its effective factors using artificial neural network: A case study of children and adolescents residing in Isfahan-Iran

Document Type : Original Article

Authors

1 Department of Mathematics, Payame Noor University, PO Box 19395-4697, Tehran, Iran

2 Associate Professor، Alborz University of Medical Sciences, Karaj, Iran

3 Professor- Isfahan University of Medical Sciences, Isfahan, Iran

Abstract

Introduction: Overweight obesity is now so widespread in the world. This study aims to use an artificial neural network modeling tool to develop a predictive model for the diagnosis of obesity in children and adolescents.
 
Material and methods: Participants consisted of 460 school students, aged 7-18 years, who studied in a national school-based surveillance program (CASPIAN-V). Training network with 10 input variables including: age, sex, weight, height, waist circumference, systolic blood pressure, diastolic blood pressure, body mass index, waist-to-height ratio, physical activity, and with output variable obesity with 17 and 15 hidden neurons for girls and boys was designed.
 
Results: After designing the network, the value of gradient on the data was 0.0021194 for girls and 0.0031658 for boys. The sensitivity, specificity and accuracy of the neural network were 0.9444, 0.9855, 0.9822, respectively in girls, and 0.9655, 0.9757, 0.9755 in boys;  in all these cases, the designed artificial neural network performed better than waist circumference and body mass index. A review of the final weights of this network showed that the input variable body mass index in girls and the input variable waist-to-height ratio in boys had the most influence in diagnosis of obesity.
 
Conclusions: Our results show that although body mass index has a better diagnostic performance in determining excess body fat than waist circumference, in boys and girls of both groups, and also in all parameters of sensitivity, specificity and accuracy, the artificial neural network acts better than body mass index and waist circumference, so that with an accuracy of more than 96%, we can detects obesity.
 
 

Keywords

References

1. Boutayeb A, Boutayeb S. The burden of non communicable diseases in developing countries. Int J Equity Health. 2005;4(1):2. doi: 10.1186/1475-9276-4-2.
2. Heshmat R, Hemati Z, Qorbani M, Nabizadeh Asl L, Motlagh ME, Ziaodini H, et al. Metabolic syndrome and associated factors in Iranian children and adolescents: the CASPIAN-V study. J Cardiovasc Thorac Res. 2018;10(4):214-20. doi: 10.15171/jcvtr.2018.37.
3. Vigna L, Brunani A, Brugnera A, Grossi E, Compare A, Tirelli AS, et al. Determinants of metabolic syndrome in obese workers: gender differences in perceived job-related stress and in psychological characteristics identified using artificial neural networks. Eat Weight Disord. 2019;24(1):73-81. doi: 10.1007/s40519-018-0536-8.
4. Kelishadi R, Hovsepian S, Djalalinia S, Jamshidi F, Qorbani M. A systematic review on the prevalence of metabolic syndrome in Iranian children and adolescents. J Res Med Sci. 2016;21:90. doi: 10.4103/1735-1995.192506.
5. Kelishadi R, Poursafa P. A review on the genetic, environmental, and lifestyle aspects of the earlylife origins of cardiovascular disease. Curr Probl Pediatr Adolesc Health Care. 2014;44(3):54-72. doi: 10.1016/j.cppeds.2013.12.005.
6. Baygi F, Tabatabaee M, Dorosty AR, Eshraghian M. The effective factors in obesity of Neishabur children. Zahedan Journal of Research in Medical Sciences. 2011;13(4):24-8. Persian.
7. Ebrahimpour P, Fakhrzadeh H, Pourebrahim R, Hamidi A, Heshmat R, Nouri M, et al. Metabolic Syndrome and its Relationship with Insulin Levels in Obese Children: A Study on Primary School Children of the 6th District of Tehran. Razi Journal of Medical Sciences. 2006;13(51):7-16.
8. Kopelman PG. Obesity as a medical problem. Nature. 2000;404(6778):635-43. doi: 10.1038/35007508.
9. Scheffler C, Schuler G. Analysis of BMI of preschool children--results of longitudinal studies. Anthropol Anz. 2009;67(1):53-63. doi: 10.1127/0003-5548/2009/0007.
10. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ. 2000;320(7244):1240-3. doi: 10.1136/bmj.320.7244.1240.
11. Worachartcheewan A, Shoombuatong W, Pidetcha P, Nopnithipat W, Prachayasittikul V, Nantasenamat C. Predicting Metabolic Syndrome Using the Random Forest Method. ScientificWorldJournal. 2015;2015:581501. doi: 10.1155/2015/581501.
12. Duran I, Martakis K, Rehberg M, Semler O, Schoenau E. Diagnostic performance of an artificial neural network to predict excess body fat in children. Pediatr Obes. 2019;14(2):e12494. doi: 10.1111/ijpo.12494.
13. Prentice AM, Jebb SA. Beyond body mass index. Obes Rev. 2001;2(3):141-7. doi: 10.1046/j.1467-789x.2001.00031.x.
14. Brambilla P, Bedogni G, Heo M, Pietrobelli A. Waist circumference-to-height ratio predicts adiposity better than body mass index in children and adolescents. Int J Obes (Lond). 2013;37(7):943-6. doi: 10.1038/ijo.2013.32.
15. Kriemler S, Puder J, Zahner L, Roth R, Meyer U, Bedogni G. Estimation of percentage body fat in 6- to 13-year-old children by skinfold thickness, body mass index and waist circumference. Br J Nutr. 2010;104(10):1565-72. doi: 10.1017/S0007114510002357.
16. Atkov OY, Gorokhova SG, Sboev AG, Generozov EV, Muraseyeva EV, Moroshkina SY, et al. Coronary heart disease diagnosis by artificial neural networks including genetic polymorphisms and clinical parameters. J Cardiol. 2012;59(2):190-4. doi: 10.1016/j.jjcc.2011.11.005.
17. Kuduvalli M, Grayson AD, Oo AY, Fabri BM, Rashid A. Risk of morbidity and in-hospital mortality in obese patients undergoing coronary artery bypass surgery. European journal of cardiothoracic surgery. 2002;22(5):787-93. doi: 10.1016/S10107940(02)00448-7.
18. Horwich TB, Fonarow GC, Hamilton MA, MacLellan WR, Woo MA, Tillisch JH. The relationship between obesity and mortality in patients with heart failure. J Am Coll Cardiol. 2001;38(3):789-95. doi: 10.1016/s0735-1097(01)01448-6.
19. Valavanis IK, Mougiakakou SG, Grimaldi KA, Nikita KS. A multifactorial analysis of obesity as CVD risk factor: use of neural network based methods in a nutrigenetics context. BMC Bioinformatics. 2010;11:453. doi: 10.1186/1471-2105-11-453.
20. Liu J, Tang ZH, Zeng F, Li Z, Zhou L. Artificial neural network models for prediction of cardiovascular autonomic dysfunction in general Chinese population. BMC Med Inform Decis Mak. 2013;13:80. doi: 10.1186/1472-6947-13-80.
21. Tang ZH, Liu J, Zeng F, Li Z, Yu X, Zhou L. Comparison of prediction model for cardiovascular autonomic dysfunction using artificial neural network and logistic regression analysis. PLoS One. 2013;8(8):e70571. doi: 10.1371/journal.pone.0070571.
22. Tahamtan RAM, Esmaeili MH, Ghaemian A, Esmaeili J. Application of artificial neural network for assessing coronary artery disease. J Mazandaran Univ Med Sci. 2012;22(86):8-17.Persian.
23. Ferenci T, Kovacs L. Predicting body fat percentage from anthropometric and laboratory measurements using artificial neural networks. Applied Soft Computing. 2018;67:834-9. doi: 10.1016/j.asoc.2017.05.063.
24. Kupusinac A, Stokic E, Doroslovacki R. Predicting body fat percentage based on gender, age and BMI by using artificial neural networks. Comput Methods Programs Biomed. 2014;113(2):610-9. doi: 10.1016/j.cmpb.2013.10.013.
25. Shao YE. Body fat percentage prediction using intelligent hybrid approaches. ScientificWorldJournal. 2014;2014:383910. doi: 10.1155/2014/383910.
26. Ergun U. The classification of obesity disease in logistic regression and neural network methods. J Med Syst. 2009;33(1):6772. doi: 10.1007/s10916-008-9165-5.
27. Heydari ST, Ayatollahi SM, Zare N. Comparison of artificial neural networks with logistic regression for detection of obesity. J Med Syst. 2012;36(4):2449-54. doi: 10.1007/s10916-011-9711-4.
28. Livingstone DJ. Artificial neural networks. Methods and applications. Preface. Methods Mol Biol. 2008;458. doi: 10.1007/9781-60327-101-1.
29. Cross SS, Harrison RF, Kennedy RL. Introduction to neural networks. Lancet. 1995;346(8982):1075-9. doi: 10.1016/s01406736(95)91746-2.
30. Dunne RA. A statistical approach to neural networks for pattern recognition. New Jersey: John Wiley & Sons; 2007. doi: 10.1002/9780470148150.
31. Bishop CM. Neural networks for pattern recognition. Oxford: Oxford university press; 1995.
32. Sivanandam S, Deepa S. Introduction to neural networks using Matlab 6.0. New York: Tata McGraw-Hill Education; 2006.
33. Kurt I, Ture M, Kurum AT. Comparing performances of logistic regression, classification and regression tree, and neural networks for predicting coronary artery disease. Expert systems with applications. 2008;34(1):366-74. doi: 10.1016/j.eswa.2006.09.004.
34. Ture M, Kurt I, Kurum AT, Ozdamar K. Comparing classification techniques for predicting essential hypertension. Expert Systems with Applications. 2005;29(3):583-8. doi: 10.1016/j.eswa.2005.04.014.
35. Menhaj M. Computational intelligence fundamentals of neural network. 2nd ed. Tehran: Amirkabir University of Technology Publication; 2008. Persian.
36. Fausett LV. Fundamentals of neural networks: architectures, algorithms, and applications. New Jersey: Prentice-Hall, Inc. Upper Saddle River; 1994.
37. Motlagh ME, Ziaodini H, Qorbani M, Taheri M, Aminaei T, Goodarzi A, et al. Methodology and Early Findings of the Fifth Survey of Childhood and Adolescence Surveillance and Prevention of Adult Noncommunicable Disease: The CASPIAN-V Study. Int J Prev Med. 2017;8:4. doi: 10.4103/2008-7802.198915.
38. Chen Y, Zhang Y, Wang L. Low Diagnostic Accuracy of Body Mass Index-Based and Waist Circumference-Based References of Childhood Overweight and Obesity in Identifying Overfat among Chinese Children and Adolescents. Biomed Res Int. 2018;2018:4570706. doi: 10.1155/2018/4570706.
39. Javed A, Jumean M, Murad MH, Okorodudu D, Kumar S, Somers VK, et al. Diagnostic performance of body mass index to identify obesity as defined by body adiposity in children and adolescents: a systematic review and metaanalysis. Pediatr Obes. 2015;10(3):234-44. doi: 10.1111/ijpo.242.
40. Gurney K. An introduction to neural networks. Boca Raton: CRC Press; 1997. doi: 10.4324/9780203451519.
41. Wohlfahrt-Veje C, Tinggaard J, Winther K, Mouritsen A, Hagen CP, Mieritz MG, et al. Body fat throughout childhood in 2647 healthy Danish children: agreement of BMI, waist circumference, skinfolds with dual X-ray absorptiometry. Eur J Clin Nutr. 2014;68(6):664-70. doi: 10.1038/ejcn.2013.282.
42. Flegal KM, Shepherd JA, Looker AC, Graubard BI, Borrud LG, Ogden CL, et al. Comparisons of percentage body fat, body mass index, waist circumference, and waist-stature ratio in adults. Am J Clin Nutr. 2009;89(2):500-8. doi: 10.3945/ajcn.2008.26847.
43. Choi DH, Hur YI, Kang JH, Kim K, Cho YG, Hong SM, et al. Usefulness of the Waist Circumference-to-Height Ratio in Screening for Obesity and Metabolic Syndrome among Korean Children and Adolescents: Korea National Health and Nutrition Examination Survey, 2010-2014. Nutrients. 2017;9(3). doi: 10.3390/nu9030256.
Health Management & Information Science
Volume 8, Issue 3
July 2021
Pages 177-184

Files

Share

How to cite

Statistics