Early Diagnosis of Diabetes Mellitus Using Machine Learning Algorithms and Clinical Data Analysis

Abstract

The catastrophic effects of diabetes affect a significant majority of individuals worldwide, and many of these cases are not detected in time. The inability of the body to create enough insulin results in diabetes, which raises blood glucose levels and has become one of the main causes of death. The impact of diabetes has increased recently and is predicted to keep expanding on a global scale. Globally, there are currently 463 million people with diabetes, and by 2045, that number is expected to increase to 700 million. Type 2 diabetes is becoming more and more common in many nations. Diabetes is linked to serious health issues such heart disease, stroke, kidney damage, and blindness.

To address these issues, this study makes use of machine learning algorithms, which process and learn from vast amounts of data, to improve diabetes prediction and early detection. In particular, this study compares the accuracy of diabetes diagnosis using the Random Forest Classifier, Decision Tree Classification, and Logistic Regression algorithms to ascertain how well they predict diabetes outcomes.