Optimised ensemble learning-based IoT-enabled heart disease monitoring system: an optimal fuzzy ranking concept

Abstract

ABSTRACT Generally, patients with heart diseases enquire for medical services while sick. They will not experience sickness often until they reach the last stage of heart disease, and so the damages are not reversible. Thus, the solution for reducing the death rate of heart disease healthcare is becoming the passive medical care type. Physicians usually monitor the physical status of patients, which helps in deciding when to provide medical services by considering the real-time analysis of patients. Moreover, a significant division of pervasive clinical care is taken as the development of practical monitoring systems. An IoT approaches are overwhelming the progression in addressing the issue of heart disease patient care since they can modify the service mode in a pervasive way. A remote monitoring model is more significant for realising pervasive healthcare services. Here, an ‘optimised ensemble learning’-based IoT-enabled heart disease monitoring system through an optimised ensemble fuzzy ranking (OEFR) strategy, deep feature extraction and heuristic improvement. Moreover, the hyperparameter of each classifier is optimised through a new improved dingo optimiser (I-DOX) algorithm. Throughout the result analysis, the accuracy of the designed I-DOX-based OEFR model is attained 96.72%. Thus, the analysis exhibits an overall performance analysis in terms of high performance.