Abstract

This paper presents an in-depth technical analysis and comparison of various machine learning models for predicting the occurrence of nephropathy in diabetic patients. The models evaluated in this study encompass a wide range of algorithms, including logistic regression, support vector machines, decision trees, random forest, naive Bayes, k-nearest neighbors, gradient boosting machines, and fully connected neural network. The performance of these models is evaluated using accuracy, precision, and recall metrics. The findings from this extensive evaluation provide valuable insights into the strengths and limitations of each model, facilitating informed decision-making for selecting the most appropriate algorithm for predicting the occurrence of nephropathy in diabetic patients. The experimental results indicated that random forest exhibited an excellent performance whereas naive bayes algorithm performed poorly.

References

• Marwa Almasoud and Tomas E Ward. Detection of chronic kidney disease using machine learning algorithms with least number of predictors. International Journal of Soft Computing and Its Applications, 10(8), 2019.
• Timothy C Au. Random forests, decision trees, and categorical predictors: the” absent levels” problem. The Journal of Machine Learning Research, 19(1):1737–1766, 2018.
• Bjorn Kaijun Betzler, Rehena Sultana, Riswana Banu, Yih Chung Tham, Cynthia Ciwei Lim, Ya Xing Wang, Vinay Nangia, E Shyong Tai, Tyler Hyungtaek Rim, Mukharram M Bikbov, et al. Association between body mass index and chronic kidney disease in asian populations: a participant- level meta-analysis. Maturitas, 154:46–54, 2021.
• D Campagna, A Alamo, A Di Pino, C Russo, AE Calogero, F Purrello, and R Polosa. Smoking and diabetes: dangerous liaisons and confusing relationships. Diabetology & metabolic syndrome, 11(1):1–12, 2019.
• Haleh Chehregosha, Mohammad E Khamseh, Mojtaba Malek, Farhad Hosseinpanah, and Faramarz Ismail-Beigi. A view beyond hba1c: role of continuous glucose monitoring. Diabetes Therapy, 10:853–863, 2019.
• Zewei Chen, Xin Zhang, and Zhuoyong Zhang. Clinical risk assessment of patients with chronic kidney disease by using clinical data and multivariate models. International urology and nephrology, 48:2069–2075, 2016.
• Dervla M Connaughton, Claire Kennedy, Shirlee Shril, Nina Mann, Susan L Murray, Patrick A Williams, Eoin Conlon, Makiko Nakayama, Amelie T van der Ven, Hadas Ityel, et al. Monogenic causes of chronic kidney disease in adults. Kidney international, 95(4):914–928, 2019.
• Tommaso Di Noia, Vito Claudio Ostuni, Francesco Pesce, Giulio Binetti, David Naso, Francesco Paolo Schena, and Eugenio Di Sciascio. An end stage kidney disease predictor based on an artificial neural networks ensemble. Expert systems with applications, 40(11):4438–4445, 2013.
• Jerome H Friedman. Greedy function approximation: a gradient boosting machine. Annals of statistics, pages 1189–1232, 2001.
• Kai Hakala, Suwisa Kaewphan, Jari Bjo¨rne, Farrokh Mehryary, Hans Moen, Martti Tolvanen, Tapio Salakoski, and Filip Ginter. Neural network and random forest models in protein function prediction. IEEE/ACM Transactions on Compu- tational Biology and Bioinformatics, 19(3):1772–1781, 2020.
• Edgar A Jaimes, Ming-Sheng Zhou, Mohammed Siddiqui, Gabriel Rezonzew, Runxia Tian, Surya V Seshan, Alecia N Muwonge, Nicholas J Wong, Evren U Azeloglu, Alessia Fornoni, et al. Nicotine, smoking, podocytes, and diabetic nephropathy. American Journal of Physiology-Renal Physiology, 320(3):F442–F453, 2021.
• Naseer Ullah Khan, Jing Lin, Xukun Liu, Haiying Li, Wei Lu, Zhuning Zhong, Huajie Zhang, Muhammad Waqas, and Lim- ing Shen. Insights into predicting diabetic nephropathy using urinary biomarkers. Biochimica et Biophysica Acta (BBA)- Proteins and Proteomics, 1868(10):140475, 2020.
• Andrzej Konieczny, Jakub Stojanowski, Magdalena Krajew- ska, and Mariusz Kusztal. Machine learning in prediction of iga nephropathy outcome: A comparative approach. Journal of Personalized Medicine, 11(4):312, 2021.
• Csaba P Kovesdy. Epidemiology of chronic kidney disease: an update 2022. Kidney International Supplements, 12(1):7– 11, 2022.
• Wei Li, Weixiang Luo, Mengyuan Li, Liyu Chen, Liyan Chen, Hua Guan, and Mengjiao Yu. The impact of recent devel- opments in electrochemical poc sensor for blood sugar care. Frontiers in Chemistry, 9:723186, 2021.
• Subramani Mani, Yukun Chen, Tom Elasy, Warren Clayton, and Joshua Denny. Type 2 diabetes risk forecasting from emr data using machine learning. In AMIA annual symposium proceedings, volume 2012, page 606. American Medical Infor- matics Association, 2012.
• Vijayakumar Natesan and Sung-Jin Kim. Diabetic nephropathy–a review of risk factors, progression, mechanism, and dietary management. Biomolecules & therapeutics, 29(4):365, 2021.
• Jiongming Qin, Lin Chen, Yuhua Liu, Chuanjun Liu, Chang- hao Feng, and Bin Chen. A machine learning methodology for diagnosing chronic kidney disease. IEEE Access, 8:20991– 21002, 2019.
• Violeta Rodriguez-Romero, Richard F Bergstrom, Brian S Decker, Gezim Lahu, Majid Vakilynejad, and Robert R Bies. Prediction of nephropathy in type 2 diabetes: an analysis of the accord trial applying machine learning techniques. Clinical and translational science, 12(5):519–528, 2019.
• S Rasoul Safavian and David Landgrebe. A survey of deci- sion tree classifier methodology. IEEE transactions on sys- tems, man, and cybernetics, 21(3):660–674, 1991.
• Hanny Sawaf, George Thomas, Jonathan J Taliercio, Georges Nakhoul, Tushar J Vachharajani, and Ali Mehdi. Therapeutic advances in diabetic nephropathy. Journal of Clinical Medicine, 11(2):378, 2022.
• Hiroe Seto, Asuka Oyama, Shuji Kitora, Hiroshi Toki, Ryohei Yamamoto, Jun’ichi Kotoku, Akihiro Haga, Maki Shinzawa, Miyae Yamakawa, Sakiko Fukui, et al. Gradient boosting decision tree becomes more reliable than logistic regression in predicting probability for diabetes with big data. Scientific Re- ports, 12(1):15889, 2022.
• Alvaro Sobrinho, Andressa CM Da S Queiroz, Leandro Dias Da Silva, Evandro De Barros Costa, Maria Eliete Pinheiro, and Angelo Perkusich. Computer-aided diagnosis of chronic kidney disease in developing countries: A comparative anal- ysis of machine learning techniques. IEEE Access, 8:25407– 25419, 2020.
• Katherine R Tuttle, Radica Z Alicic, O Kenrik Duru, Cami R Jones, Kenn B Daratha, Susanne B Nicholas, Sterling M McPherson, Joshua J Neumiller, Douglas S Bell, Carol M Mangione, et al. Clinical characteristics of and risk fac- tors for chronic kidney disease among adults and children: an analysis of the cure-ckd registry. JAMA network open, 2(12):e1918169–e1918169, 2019.
• Ginny Y Wong, Frank HF Leung, and Sai-Ho Ling. Predict- ing protein-ligand binding site using support vector machine with protein properties. IEEE/ACM Transactions on Computational Biology and Bioinformatics, 10(6):1517–1529, 2013.