Enhanced disease diagnosis through edge computing with deep learning for advanced medical analysis

Abstract

Edge computing (EC) has appeared as a significant tool due to increasing number of information as well as expanding amount of its requirements utilized for processing the data. This technology is implemented on a source terminal, it collects and processes the information and grants near-end services to the user. In recent years, the modern environment has presented elegant healthcare monitoring schemes for the earlier prediction and identification of various diseases. However, the information sharing from one location to another is quietly time-consuming and necessitates a significant amount of energy, EC gives the best solution for rectifying this kind of hazard, EC is widely incorporated with various learning schemes for disease diagnosis. EC offers intelligence based Internet of Things (IoT) enabled stable services for securing the data and systems. This survey concerns two major contributions, in the first contribution presents an intelligent heart disease prediction method, which collects individual data from gadgets. This information is gathered from the hardware components like sensors deployed in human blood. The gateway components collect information from various nodes and transfer it to the sub-nodes for the detection of cardiac disease. Sub-nodes process signals like amplitude, heartbeat, mean, variance, and energy with significant features that are derived in parallel. In addition, the variability of signals is a computer for eliminating the redundant characteristics. At last, the recovered features are transferred to the developed diagnostic model, which is employed with the Convolutional Neural Network based Sun Flower Optimization (CNN-optimized SFO) technique to predict whether the individual is sick or not newline