Designing a Fuzzy Expert System for Diagnosis and Prediction of Metabolic Syndrome in Children and Adolescents

Document Type : Original Article

Authors

1 Assistant Professor of Applied Mathematics, Payame Noor University, Tehran, Iran

2 PhD Student in Applied Mathematics, Payame Noor University, Tehran, Iran

Abstract

Introduction: Metabolic Syndrome (MetS) is one of the most common metabolic disorders
seen in children and adolescents. In this study, the prevalence of MetS and its related factors
are evaluated using a fuzzy expert system (FES) in a national representative sample of age
groups.
Methods: The FES is designed based on the data of 800 participants of the fifth study of the
program for monitoring and prevention of non-communicable diseases among children and
adolescents in Iran in 2015. The data of 560 participants were used as training data and 240 as
test data were used to test the rules and output of the system. The fuzzy system that has been
designed includes input data (age, waist, systolic blood pressure, diastolic blood pressure,
BMI, waist-to-height ratio, nutrition, and abdominal obesity), and at the end gives us an
output that diagnoses the health status with MetS or predicts the disease.
Results: The analysis shows that this method, with an accuracy of more than 98%, can predict
and diagnose MetS among children and adolescents better than other methods.
Conclusion: The fuzzy system is designed to accept multiple variables simultaneously as
input variables and also use more people information than similar research as primary data.
In addition, its accuracy is more than 98%. Preliminary data were collected from children and
adolescents with different lifestyles across the country. This system can act as an assistant in
the service of a specialist doctor to diagnose the disease.

Keywords

References

1. Dehghandar M, Hassani Bafrani A, Dadkhah M, Qorbani M, Kelishadi R. Diagnosis of Obesity and Hypertension in Isfahani Students Using Artificial Neural Network. Journal of Health and Biomedical Informatics. 2021;8(1):12-23.
2. Dehghandar M, Pabasteh M, Heydari R. Diagnosis of COVID-19 disease by fuzzy expert system designed based on input-output. Journal of Control. 2021;14(5):71-78. Persian.
3. Sheikh Taheri A, Hamedan F, Sandgol H, Orouji A. Establishment of a fuzzy expert system for diagnosing chronic kidney disease. Razi Journal of Medical Sciences. 2017;25(10).
4. Comprehensive database of health education and culture. [cited 4 October 2012]. Available rom: http://marizkhone.com/2012/11/04. Persian.
5. Kaplan NM. The deadly quartet. Upperbody obesity, glucose intolerance, hypertriglyceridemia, and hypertension. Arch Intern Med. 1989;149(7):1514-20. doi: 10.1001/ archinte.149.7.1514.
6. Deepa M, Farooq S, Datta M, Deepa R, Mohan V. Prevalence of metabolic syndrome using WHO, ATPIII and IDF definitions in Asian Indians: the Chennai Urban Rural Epidemiology Study (CURES-34). Diabetes Metab Res Rev. 2007;23(2):127-34. doi: 10.1002/dmrr.658.
7. Zarkesh M, Daneshpour MS, Faam B, Fallah MS, Hosseinzadeh N, Guity K, et al. Heritability of the metabolic syndrome and its components in the Tehran Lipid and Glucose Study (TLGS). Genet Res (Camb). 2012;94(6):331-7. doi: 10.1017/ S001667231200050X.
8. Langarizade M, Khajehpour E, Khajehpour H, Noori T. Fuzzy Expert System for Diagnosis of Bacterial Meningitis from Other Types of Meningitis in Children. Journal of Health and Biomedical Informatics. 2014;1(1):19-25.
9. Grundy SM. Metabolic syndrome pandemic. Arterioscler Thromb Vasc Biol. 2008;28(4):629-36.
10. Zimmet P, Alberti G, Kaufman F, Tajima N, Silink M, Arslanian S, et al. The metabolic syndrome in children and adolescents. Lancet. 2007;369(9579):2059-61. doi: 10.1016/S0140-6736(07)60958-1.
11. Karimian Z, Kojouri J, Sagheb MM. A Review of the Evidence-Based Medicine Realm Based on Two Factors: The Nature of Science and Decision Making Situation. Interdisciplinary Journal of Virtual Learning in Medical Sciences. 2020;6(2):69-75.
12. Cameron AJ, Shaw JE, Zimmet PZ. The metabolic syndrome: prevalence in worldwide populations. Endocrinol Metab Clin North Am. 2004;33(2):351-75, table of contents. doi: 10.1016/j. ecl.2004.03.005.
13. Salari M, Kiwani O, Soleimani A [Internet]. Designing a fuzzy expert system to diagnose the risk of hypertension. Available from: https://civilica.com/doc/219663/. Persian.
14. Korenevskiy N. Application of fuzzy logic for decision-making in medical expert systems. Biomedical engineering. 2015;49(1):46-9. doi: 10.1007/s10527-015-9494-x.
15. Khosravi M, Zarei H, Kheiri Donighi S; Diabetes based on fuzzy systems and Wall optimization algorithm. Journal of New Ideas in Science, Engineering and Technology. 2018;2(2):62-75. Persian.
16. Sedehi M, Mehrabi Y, Kazemnejad A, Hadaegh F. Comparison of artificial neural network, logistic regression and discriminant analysis methods in prediction of metabolic syndrome. Iranian journal of endocrinology and metabolism. 2010;11(6):638-46.
17. Zabah A, Eskandi Z, Sardari A, Noghandi A. Diagnosis of diabetes using artificial and neuralfuzzy neural network. Journal of Torbat-Heydariyeh University of Medical Sciences. 2018;6(2):11-9.
18. Khosravanian A, Ayat S. A Physician Assistant Intelligence System Based on Artificial Neural Network for Diabetes Diagnosis. Iranian Journal of Diabetes and Metabolism. 2019;18(2):71-9.
19. Ghaemi N, Afzal Aghaee M. Prevalence of Metabolic Syndrome in children with overweight. Medical Journal of Mashhad University of Medical Sciences. 2010;53(2):98-103.
20. 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.
21. Mehrdad M, Hosseinpanah F, Azizi F. Prevalence of metabolic syndrome among 3-9 years old children inTehran Lipid and Glucose Study. Research in Medicine. 2006;30(4):337-47.
22. Ghargerechi R, Azar MR. Prevalence of Metabolic Syndrome in Obese Children and Adolescents Dolescents. Medical Journal of Tabriz University of Medical Sciences. 2010;32(3):57-61.
23. Klishadi R. Comparison of the effects of vitamin D versus placebo on the components of metabolic syndrome in children and adolescents aged 10-16. Pediatric Developmental Research Center, Isfahan
University of Medical Sciences. 2012.
24. Zadeh LA. Fuzzy sets as a basis for a theory of possibility. Fuzzy sets and systems. 1978;1(1):3-28. doi: 10.1016/01650114(78)90029-5.
25. Zadeh LA. A theory of approximate reasoning. Machine intelligence. New York: John Wiley and Sons; 1979. p. 149-94.
26. Burillo P, Bustince H, Mohedano V. Some Definitions of Intuitionistic Fuzzy Number. In: Lakov D, editor. First properties. Proceedings of the 1st workshop on fuzzy based expert systems. 1994.
27. Sharghi S. Metabolic Syndrome and Obesity. Iranian Journal of Diabetes and Lipid Disorders. 2013;12(5):399-412. Persian.
28. Saeidi M. Metabolic syndrome and hypertension in diabetic patients. Iranian Journal of Endocrinology and Metabolism. 2009;11(1):11-6.
29. Nizami F, Farooqui MS, Munir SM, Rizvi TJ. Effect of fiber bread on the management of diabetes mellitus. J Coll Physicians Surg Pak. 2004;14(11):673-6. doi: 11.2004/JCPSP.673676.
30. Su CT, Yang CH, Hsu KH, Chiu WK. Data mining for the diagnosis of type II diabetes from threedimensional body surface anthropometrical scanning data. Computers & mathematics with applications. 2006;51(6-7):1075-92. doi: 10.1016/j. camwa.2005.08.034.
31. Barakat NH, Bradley AP, Barakat MN. Intelligible support vector machines for diagnosis of diabetes mellitus. IEEE Trans Inf Technol Biomed. 2010;14(4):1114-20. doi: 10.1109/TITB.2009.2039485.
32. Worachartcheewan A, Nantasenamat C, Isarankura-Na-Ayudhya C, Pidetcha P, Prachayasittikul V. Identification of metabolic syndrome using decision tree analysis. Diabetes Res Clin Pract. 2010;90(1):e15-8. doi: 10.1016/j.diabres.2010.06.009.
33. Bagherzdeh Khiabani F, Akhavan Niaki S, editors. An Ensemble Model for Prediction of Diabetes Type 2. Tehran: The Seventh Iran Data Mining Conference/IDMC; 2013.
34. Smith JW, Everhart JE, Dickson W, Knowler WC, Johannes RS, editors. Using the ADAP learning algorithm to forecast the onset of diabetes mellitus. Proc Annu Symp Comput Appl Med Care. 1988:261-5.
35. Asrardel M. Prediction of combustion dynamics in an experimental turbulent swirl stabilized combustor with secondary fuel injection. University of Tehran. 2015.
Health Management & Information Science
Volume 8, Issue 2
April 2021
Pages 79-89

Files

Share

How to cite

Statistics