Optimized Diabetes Prediction Using Soft Computing: CURE-ADASYN for Imbalance and Advanced Deep Learning Classification Methods

Abstract

Diabetes has become a major global health concern, leading to a number of catastrophic effects such as cardiovascular problems, kidney disease, and vision loss. Deep learning algorithms have shown potential in medical services for precise disease detection and treatment, which will ease the burden on medical personnel. Rapid advancements in diabetes forecasting have opened up new avenues for patient empowerment and early intervention. In order to do this, this research suggests a novel diabetes prediction model that uses an improved LSTM classifier, feature selection using Grey Wolf Optimization, and Particle Swarm Optimization. Using performance metrics including accuracy, precision, recall, and F1 score, our method is thoroughly assessed.