Prediction of Insulin Doses for Diabetes Patients in Hospital using Voting Classifier
Article Sidebar
Main Article Content
Abstract
The diabetes patients in the hospital need to supply accurate insulin dose levels. This paper uses the 93 KB diabetes dataset, which consists of the 101766 diabetes patients. The dataset is transformed into clean data using the data preprocessing techniques. The data is split into 80-20 train and test data. The four voting classifier models, such as two hard voting and two soft voting classifier models, are developed by using random forest, K nearest neighbors, extreme gradient boosting, and logistic regression. The data is trained into the four models, and the four models’ performance is analyzed using classification reports in metrics such as precision, recall, F1-score, and accuracy. According to 0.77 precision, 0.79 recall, 0.78 f1-score, and 0.79 accuracy, the Soft Voting Classifier model SV2 is greater than the other three models, HV1, HV2, and SV1. According to accuracy, the SV2 model is the best model to predict insulin dose levels.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
References
[1] T. Poosakaew, P. Kesssomboon and J. F. Smith, “Risk Factors for Hospitalization Due to Hypoglycemia in Diabetic Patients in Northeast Thailand,” J. Diabet. Mellit., vol. 4, pp. 165-171, 2014.
[2] O. A. Folasade, N. Yekini and R. Babatunde, “Classification of Diabetes Types using Machine Learning,” Int. J. Adv. Comput. Sci. and Appl., vol. 13, no. 9, pp. 152-161, 2022.
[3] T. Singye and S. Unhapipat, “Time Series Analysis of Diabetes Patients: A Case Study of Jigme Dorji Wangchuk National Referral Hospital in Bhutan,” IOP Conf. Ser.: J. Phys., vol. 1039, no. 1, pp. 1–11, 2018.
[4] Diabetes in Hospital: Health Care Providers’ Guide to Diabetes Management for Inpatients, National Diabetes Services Scheme in Australian, Ver. 2, 2022.
[5] J. G. Mosquera, M. V. Monedero, M. G. Martin and R. S. Segundo, “Transformer-Based Prediction of Hospital Readmissions for Diabetes Patients,” Electron., vol. 14, no. 1, pp. 1-24, 2025.
[6] P. Symeonidis, E. C. Rizos, D. Sotiriou and M. Zanker, “Predicting the Optimal Insulin Dose and Controlling the Blood Glucose Levels with Deep Reinforcement Learning,” Res. Square, PPR917982, pp. 1-26, 2024.
[7] N. A. Noori and A. A. Yassin, “A Comparative Analysis for Diabetic Prediction Based on Machine Learning Techniques,” J. Barah Res. (Sci.), vol. 47, no. 1, pp. 180-190, 2021.
[8] Y. A. Ozturkmen et al., “The Effect of Type and Daily Doses of Insulin to Treatment Success in Type 2 Diabetes Patients Who are Receiving Basal-bolus Insulin Therapy,” The Med. Bullet. of Sisli Etfal Hosp., vol. 54, no. 4, pp. 416-423, 2020.
[9] A. Jafar, A. Kobayati, M. A. Tsoukas and A. Haidar, “Personalized Insulin Dosing using Reinforcement Learning for High-Fat Meals and Aerobic Exercises in Type 1 Diabetes: A Proof-of-Concept Trial,” Nat. Commun., vol. 15, no.1, pp. 6585, 2024.
[10] P. Symeonidis et al, “Deep Reinforcement Learning for Personalized Insulin Dosing and Glucose Control of Hospitalized in ICU Patients,” Int. J. Data Sci. and Analy., vol. 20, no. 7, pp. 6841-6854, 2025.
[11] B. Silver et al., “EADSG Guidelines: Insulin Therapy in Diabetes,” Diabet. Ther., vol. 9, pp. 449-492, 2018.
[12] B. W. Lee et al., “Insulin Therapy for Adult Patients with Type 2 Diabetes Mellitus: A Position Statement of the Korean Diabetes Association, 2017,” Diabet. & Metabol. J., vol. 41, no. 5, pp. 367-373, 2017.
[13] K. Gupta and N. Jiwani, “Prediction of Insulin Level of Diabetes Patient Using Machine Learning Approaches,” Int. J. Creat. Res. Thoughts, vol. 10, no. 8, pp. 434-441, 2022.
[14] V. K. Yata et al., “Development of A Machine Learning-Based Interface for Insulin Dependency Prediction Using Clinical Data,” J. Sci. Rep., vol. 15, no. 1, pp. 38720, 2025.
[15] X. Liu, B. S. I. Jankovic and J. H. Chen, “Predicting Inpatient Glucose Levels and Insulin Dosing by Machine Learning on Electronic Health Records,” MedRxiv, PPR116292, 2020.
[16] S. J. Chu, N. Amarasiri, S. Giri and P. Kafle, “Blood Glucose Level Prediction in Type 1 Diabetes Using Machine Learning,” arXiv:2502.00065, 2025.