Integrating Smart IoT and AI-Enhanced Systems for Predictive Diagnostics Disease in Healthcare

A developing technology known as the IoT allows for the connectivity of seemingly innocuous gadgets to the web. The IoT has linked millions of sensors and smart devices, which has transformed the healthcare sector. IoT devices driven by AI can identify diabetes by continuously gathering accurate glucose values. This study addresses the critical need for early and accurate diabetes detection using advanced ML algorithms, leveraging the Pima Indian Diabetes Dataset. An accuracy of diagnosing diabetes and non-diabetic cases is evaluated and compared in this study using models such as CNN, XG-Boost, DT, and SVM. Show that the model may improve diagnostic accuracy and facilitate early intervention by conducting a thorough performance evaluation utilising confusion matrices and measures like recall, accuracy, precision, and F1-score. CNN demonstrates superior accuracy of 99% with high precision of 98% and recall rates of 99.9%, positioning it as the most effective model for diabetes detection in this study. The findings underscore the feasibility of DL approaches in healthcare data analysis, contributing to improved healthcare outcomes in diabetes diagnostics.