Autonomic edge cloud assisted framework for heart disease prediction using RF-LRG algorithm.

Due to changing lifestyles, human physical exercise has dropped rapidly which leads to several health-related problems. It is not easy to detect heart-related diseases, but with the emergence and advancement in technologies like Edge, Fog computing, Machine learning, Cloud computing, and the Internet of Things (IoT), it's a cinch to track heart diseases. Cloud computing provides the resources for computation as well as online storage over the internet, but it does not support latency-sensitive and real-time applications. To overcome this bottleneck, a new emerging technology named Edge Computing is used to bring the computation resource to local nodes for the services and reduces the latency as compared to the cloud. Current Edge computing-based models failed to achieve the output of real-time applications along with high accuracy and low latency simultaneously. Hence, we have proposed an autonomic Edge-assisted Cloud-IoT framework for smart healthcare that used a Random Forest and Logistic Regression Grid (RF-LRG) approach at edge nodes for analysis of heart disease and improves the various influential parameters such as accuracy (3.88%, 7.66% and 14.18%), precision (3.7%, 9%, and 16.6%), F1 Score (5%, 7.7%, and 16.9%.), recall (3.7%, 10.5%, and 15.06%.) compared with LR, RF and KNN algorithm. The simulation results ensured that the proposed framework using the RF-LRG algorithm predicted and diagnosed heart diseases with more accuracy and reduced the latency and energy consumption significantly when compared between the cloud and edge paradigms. [ABSTRACT FROM AUTHOR]