Next-Gen Public Healthcare: Modeling an Advanced Framework for Diabetes Management

Abstract

Advancements in biotechnology and public health technologies have significantly increased public health data, aiding early disease detection and prevention, particularly in diabetes, which can lead to serious public health issues. In this study, we propose a novel crow search-driven dynamic random forest for (CS-DRF) for diabetes management reorganization. The goal of this integrated strategy is to enhance diabetes treatment results and diagnosis. The Pima Indian Diabetes (PID) dataset is gathered from open-source Kaggle website for diabetes management. We implement our recommended evaluation technique using Python software. The findings showed that the CS-DRF performed more effectively than the others regarding F1-score-93.25%, accuracy-95.46%, specificity-92.38 %, sensitivity-92.22%, and precision-93.28%. The study's validation results demonstrate how advanced the structure model based on machine learning is in detecting diabetes management.