Improving Diabetic Patients Monitoring System Using (NCA-CNN) Algorithm based on loT

Abstract

The Internet of Things (IoT) and Artificial Intelligence (AI), particularly Machine Learning (ML), have both seen significant advancements in recent years, which has resulted in significant leaps forward in the development of health monitoring systems. Patients may now be prevented, diagnosed, and monitored remotely and at home, eliminating the need to go to health and treatment centers or spend a significant amount of money doing so. This is made possible by advancements in technology. Deep learning has been the primary focus of this research as it relates to the development of a remote health monitoring system for the diagnosis of diabetes. In the system that has been suggested, improvements have been made to both the precision of the detection and the swiftness of the data processing. The Neighbourhood Component Analysis-Convolutional Neural Network (NCA-CNN) approach that we have presented involves two stages: the first stage involves picking the most important features from all of the data, and the second stage involves categorizing the chosen features. The NCA algorithm is a mathematical method that rates the characteristics based on the results of an analysis of the data and picks the most significant aspects. After that, the most salient characteristics are categorized by a deep convolutional neural network, and an accurate diagnosis of the condition is accomplished. According to the findings that were collected, the accuracy of the approach that was suggested is 97.12%.