Comparison of Performances of Machine Learning Algorithms for Predicting Diabetes Mellitus in Early Stage

Abstract

Diabetes mellitus is one of the harmful diseases that cause abnormalities in blood sugar. Early diagnosis of this disease prevents organ deterioration that may occur in the human body. Artificial intelligence-based studies are carried out effectively in the medical field. Computer-aided expert systems based on machine learning are quite useful for the early detection of this disease. The diabetes mellitus problem in this study is a classical supervised binary classification problem. There are 16 attributes in this dataset also it includes a total of 520 samples, 200 of which are negative samples and 320 of which are positive samples. The performances of the models constructed using Random Forest, Gradient Boosting, K-Nearest Neighbour, Deep Neural Networks, and finally voting ensemble classifier are analyzed within the framework of hold-out and 5-fold cross-validation techniques on the dataset pre-processed. In both scenarios, the Voting ensemble classifier presented the best performance in experiments. Accordingly, the Voting ensemble classifier offered a classification accuracy of 100% in experiments with the hold out technique, and an average of 97.31% in experiments with 5-fold cross-validation. As a result, an expert system that early diagnoses diabetes in real-time can be designed by using the Voting ensemble classifier.

Keywords

• Diabetes mellitus
• machine learning
• voting ensemble classifier.

Diyabet Hastalığının Erken Aşamada Tahmin Edilmesi İçin Makine Öğrenme Algoritmalarının Performanslarının Karşılaştırılması

Abstract

Şeker hastalığı, kan şekerinde anormalliklere neden olan zararlı hastalıklardan biridir. Bu hastalığın erken teşhisi insan vücudunda oluşabilecek organ bozulmalarını engeller. Yapay zekâ tabanlı çalışmalar medikal alanda etkin bir şekilde gerçekleştirilmektedir. Makine öğrenmesine dayalı bilgisayar destekli uzman sistemler bu hastalığın erken teşhisi için oldukça faydalıdır. Bu çalışmadaki şeker hastalığı problemi, klasik bir denetimli ikili sınıflandırma problemidir. Bu verisetinde 16 öznitelik bulunmakta olup, 200'ü negatif örnek ve 320'si pozitif örnek olmak üzere toplam 520 örnek içermektedir. Önişlemden geçirilen veriseti üzerinde Rastgele Orman, Gradyan Arttırma, K-En Yakın Komşu, Derin Sinir Ağları ve son olarak da Oylama topluluk sınıflandırıcısı kullanılarak inşa edilen modellerin performansları dışarıda tutma ve 5-kat çapraz doğrulama senaryoları çerçevesinde analiz edilmiştir. Her iki senaryoda da, Oylama topluluğu sınıflandırıcısı, deneylerde en iyi performansı sundu. Buna göre, Oylama topluluğu sınıflandırıcısı, tutma tekniğiyle yapılan deneylerde %100'lük bir sınıflandırma doğruluğu ve 5 kat çapraz doğrulamalı deneylerde ortalama %97,31'lik bir sınıflandırma doğruluğu sundu. Sonuç olarak, Oylama topluluğu sınıflandırıcısı kullanılarak diyabeti gerçek zamanlı olarak erken teşhis eden bir uzman sistem tasarlanabilir.

Keywords

• Şeker hastalığı
• makine öğrenmesi
• oylama topluluk sınıflandırıcısı

References

1. [1] S. Kumari, D. Kumar, and M. Mittal, “An ensemble approach for classification and prediction of diabetes mellitus using soft voting classifier,” International Journal of Cognitive Computing in Engineering, vol. 2, pp. 40–46, 2021.
2. [2] M. Alehegn and R. Joshi, “Analysis and prediction of diabetes diseases using machine learning algorithm: Ensemble approach,” International Research Journal of Engineering and Technology, vol. 4, no.10, pp. 426-436, 2017.
3. [3] A. Adler et al., “Reprint of: Classification of Diabetes Mellitus,” Diabetes Research and Clinical Practice, vol. 0, no. 0, p. 108972, In Press, 2021.
4. [4] R. D. Howsalya Devi, A. Bai, and N. Nagarajan, “A novel hybrid approach for diagnosing diabetes mellitus using farthest first and support vector machine algorithms,” Obesity Medicine, vol. 17, p. 100152, 2020.
5. [5] P. Zimmet, K. G. M. M. Alberti, and J. Shaw, “Global and societal implications of the diabetes epidemic,” Nature, vol. 414, no. 6865, pp. 782–787, 2001.
6. [6] J.M. Ekoe, “Diagnosis and Classification of Diabetes Mellitus,” Encyclopedia of Endocrine Diseases (Second Edition), vol. 1, pp. 105–109, 2019.
7. [7] M. Maniruzzaman et al., “Comparative approaches for classification of diabetes mellitus data: Machine learning paradigm,” Computer Methods and Programs in Biomedicine, vol. 152, pp. 23–34, 2017.
8. [8] F. Mercaldo, V. Nardone, and A. Santone, “Diabetes Mellitus Affected Patients Classification and Diagnosis through Machine Learning Techniques,” Procedia Computer Science, vol. 112, pp. 2519–2528, 2017.

9. [9] W. H. O. E. C. on Diabetes Mellitus and W. H. Organization, “Diabetes mellitus : report of a WHO Expert Committee [meeting held in Geneva from 24 to 30 November 1964].” World Health Organization, p. ger published by: Munich : Medizinische Poliklinik, 1965.
10. [10] P. Bala Manoj Kumar, R. Srinivasa Perumal, R. K. Nadesh, K. Arivuselvan, “Type 2: Diabetes mellitus prediction using Deep Neural Networks classifier,” International Journal of Cognitive Computing in Engineering, vol. 1, pp. 55–61, 2020.
11. [11] D. Jashwanth Reddy, B. Mounika, S. Sindhu, T. Pranayteja Reddy, N. Sagar Reddy, G. Jyothsna Sri, et al., “Predictive machine learning model for early detection and analysis of diabetes,” Materials Today: Proceedings, 2020.
12. [12] J. J. Khanam and S. Y. Foo, “A comparison of machine learning algorithms for diabetes prediction,” ICT Express, In Press, 2021.
13. [13] A. Viloria, Y. Herazo-Beltran, D. Cabrera, and O. B. Pineda, “Diabetes Diagnostic Prediction Using Vector Support Machines,” Procedia Computer Science, vol. 170, pp. 376–381, Jan. 2020.
14. [14] J. Chaki, S. Thillai Ganesh, S. K. Cidham, and S. Ananda Theertan, “Machine learning and artificial intelligence based Diabetes Mellitus detection and self-management: A systematic review,” Journal of King Saud University - Computer and Information Sciences, In Press, 2020.
15. [15] N. Sharma and A. Singh, “Diabetes Detection and Prediction Using Machine Learning/IoT: A Survey,". In: Luhach A., Singh D., Hsiung PA., Hawari K., Lingras P., Singh P. (eds) Advanced Informatics for Computing Research. Communications in Computer and Information Science, vol 955. Springer, Singapore, pp 471-479, 2019.
16. [16] S. Afzali and O. Yildiz, “An Effective Sample Preparation Method for Diabetes Prediction,” International Arab Journal of Information Technology, vol. 15, no. 6, 2018.
17. [17] N. Theera-Umpon, I. Poonkasem, S. Auephanwiriyakul, and D. Patikulsila, “Hard exudate detection in retinal fundus images using supervised learning", Neural Computing and Applications, vol. 32, pp. 13079–13096, 2020.
18. [18] F. Alaa Khaleel and A. M. Al-Bakry, “Diagnosis of diabetes using machine learning algorithms,” Materials Today: Proceedings, In Press, 2021.
19. [19] A. Prabha, J. Yadav, A. Rani, and V. Singh, “Design of intelligent diabetes mellitus detection system using hybrid feature selection based XGBoost classifier,” Computers in Biology and Medicine, vol. 136, pp. 104664, 2021.
20. [20] Q. Zou, K. Qu, Y. Luo, D. Yin, Y. Ju, and H. Tang, “Predicting Diabetes Mellitus With Machine Learning Techniques,” Frontiers in Genetics, vol. 9, Article 515, 2018.
21. [21] H. Lai, H. Huang, K. Keshavjee, A. Guergachi, and X. Gao, “Predictive models for diabetes mellitus using machine learning techniques,” BMC Endocrine Disorders, vol. 19, Article number: 101, pp. 1–9, Oct. 2019.
22. [22] S. NAHZAT and M. YAĞANOĞLU, “Diabetes Prediction Using Machine Learning Classification Algorithms,” Avrupa Bilim ve Teknoloji Dergisi, vol. 24, no. 24, pp. 53–59, 2021.
23. [23] N. Sneha and T. Gangil, “Analysis of diabetes mellitus for early prediction using optimal features selection,” Journal of Big Data, vol. 6, Article number: 13, pp. 1–19, 2019.
24. [24] M. M. F. Islam, R. Ferdousi, S. Rahman, and H. Y. Bushra, “Likelihood Prediction of Diabetes at Early Stage Using Data Mining Techniques,” In: Gupta M., Konar D., Bhattacharyya S., Biswas S. (eds) Computer Vision and Machine Intelligence in Medical Image Analysis. Advances in Intelligent Systems and Computing, vol. 992. pp. 113–125, Springer, Singapore, 2020.
25. [25] S. Georganos et al., “Geographical random forests: a spatial extension of the random forest algorithm to address spatial heterogeneity in remote sensing and population modelling,” Geocarto International, vol. 36, no. 2, pp. 121–136, 2019.
26. [26] N. Farnaaz and M. A. Jabbar, “Random Forest Modeling for Network Intrusion Detection System,” Procedia Computer Science, vol. 89, pp. 213–217, 2016.
27. [27] N. Aziz, E. A. P. Akhir, I. A. Aziz, J. Jaafar, M. H. Hasan, and A. N. C. Abas, “A Study on Gradient Boosting Algorithms for Development of AI Monitoring and Prediction Systems,” 2020 International Conference on Computational Intelligence, pp. 11–16, Oct. 2020.
28. [28] F. Bulut, “Obezite Riski Altındaki Çocukların Örnek Tabanlı Sınıflandırıcı Topluluklarıyla Tespiti,” Gazi Üniversitesi Mühendislik-Mimarlık Fakültesi Dergisi, vol. 32, no. 1, pp. 65–76, 2017.
29. [29] C. Sitawarin and D. Wagner, “On the Robustness of Deep K-Nearest Neighbors,” arXiv:1903.08333, 2019.
30. [30] S. Feng, H. Zhou, and H. Dong, “Using deep neural network with small dataset to predict material defects,” Materials & Design, vol. 162, pp. 300–310, Jan. 2019.
31. [31] A. Karaci, “Predicting Breast Cancer with Deep Neural Networks,” In: Hemanth D., Kose U. (eds) Artificial Intelligence and Applied Mathematics in Engineering Problems. Lecture Notes on Data Engineering and Communications Technologies, vol. 43. pp. 996–1003, Springer, Cham, 2019.
32. [32] G. Bilgin, “Investigation of The Risk of Diabetes in Early Period using Machine Learning Algorithms,” Journal of Intelligent Systems: Theory and Applications, vol. 4, no. 1, pp. 55–64, 2021.
33. [33] A. Karaci, O. Ozkaraca, E. Acar, and A. Demir, “Prediction of traumatic pathology by classifying thorax trauma using a hybrid method for emergency services,” IET Signal Processing, vol. 14, no. 10, pp. 754–764, 2020.
34. [34] C. Qi and X. Tang, “A hybrid ensemble method for improved prediction of slope stability,” International Journal for Numerical and Analytical Methods in Geomechanics, vol. 42, no. 15, pp. 1823–1839, 2018.
35. [35] O. Sagi and L. Rokach, “Ensemble learning: A survey,” Wiley Interdisciplinary Reviews-Data Mining and Knowledge Discovery, vol. 8, no. 4, pp. e1249, 2018.
36. [36] F. Moreno-Seco, J. M. Iñesta, P. J. P. de León, and L. Micó, “Comparison of Classifier Fusion Methods for Classification in Pattern Recognition Tasks,” In: Yeung DY., Kwok J.T., Fred A., Roli F., de Ridder D. (eds) Structural, Syntactic, and Statistical Pattern Recognition. SSPR /SPR 2006. Lecture Notes in Computer Science, vol. 4109. pp. 705–713, Springer, Berlin, Heidelberg, 2006.
37. [37] D.G. Altman , J.M. Bland, "Diagnostic tests. 1: Sensitivity and specificity," BMJ 1994;308:1552. https://doi.org/10.1136/BMJ.308.6943.1552.
38. [38] P. S. Kumar, K. Anisha Kumari, S. Mohapatra, B. Naik, J. Nayak, and M. Mishra, “CatBoost ensemble approach for diabetes risk prediction at early stages,” 2021 1st Odisha International Conference on Electrical Power Engineering, Communication and Computing Technology, pp. 1-6, 2021.
39. [39] T. M. Le, T. M. Vo, T. N. Pham, and S. V. T. Dao, “A Novel Wrapper-Based Feature Selection for Early Diabetes Prediction Enhanced with a Metaheuristic,” IEEE Access, vol. 9, pp. 7869–7884, 2021.
40. [40] H. N. K. Al-Behadili and K. R. Ku-Mahamud, “Fuzzy Unordered Rule Using Greedy Hill Climbing Feature Selection Method: An Application To Diabetes Classification,” Journal of Information and Communication Technology, vol. 20, no. 3, pp. 391–422, 2021.
41. [41] İ. Özer, “Uzun Kısa Dönem Bellek Ağlarını Kullanarak Erken Aşama Diyabet Tahmini," Mühendislik Bilimleri ve Araştırmaları Dergisi, vol. 2, no. 2, pp. 50–57, 2020.
42. [42] L. Chaves and G. Marques, “Data Mining Techniques for Early Diagnosis of Diabetes: A Comparative Study,” Applied Sciences, vol. 11, no. 5, pp. 1-12, 2021.