Transforming Diabetes Care with Predictive Machine Learning Techniques

• Published in: 2024 International Conference on Power, Energy, Control and Transmission Systems (ICPECTS) pp. 1 - 6
• Main Authors: Mariammal, G., Jasmine, R. Megiba, Rama Lingham N, Siva, E, Poongothai, Prema, S., S, Sai Charan
• Format: Conference Proceeding
• Language: English
• Published: IEEE 08.10.2024
• Subjects: Decision Tree; Diabetes; Logistic regression; Machine learning algorithms; Neural Networks; Numerical models; Prediction algorithms; Random Forest; Random forests; Regression tree analysis; Support Vector Machine; Support vector machines; Testing; Vectors
• DOI: 10.1109/ICPECTS62210.2024.10780095

Diabetes mellitus is a chronic metabolic disorder characterized by elevated blood sugar levels, affecting millions of individuals worldwide. Early detection and management of diabetes are crucial in preventing complications and improving patient outcomes. In recent years, machine learning techniques have shown promise in predicting diabetes risk based on various factors such as demographic information, medical history, and lifestyle habits. Verticaling through a not-too-well designed case study - This paper is an extensive examination and comparison of different ML techniques for diabetes prediction. It goes through the methodologies, datasets, features, performance metrics and challenges of current models that are based on machine learning. In addition, understand strengths and weaknesses of all the algorithms- SVM, Decision Trees, Random Forests, Logistic regression etc. Here the logistic regression algorithm is used to develop a predictive model for diabetes based on demographic and clinical features. After preprocessing the data, including handling missing values and scaling numerical features, trained a logistic regression model using a portion of the dataset. The model achieved a high accuracy of 90% on the testing set, indicating its effectiveness in distinguishing be- tween diabetic and non-diabetic individuals.