An intelligent framework for heart disease prediction deep learning-based ensemble Method

Recently, wearable sensors used in Body Area Networks (BANs) have more competencies for sensing the environments, data storage, processing, and information transfer. BANs furnish different techniques to monitor activities in various medical field applications to accurately detect heart disease. Forgiving efficient treatment for heart disease to heart patients, exact prediction is more important in medical research. A machine learning model over health care data is an important goal for heart disease prediction. Different machine learning techniques have been used in existing research that pointed out inaccurate decision-making over clinical data obtained; some improvements are needed to predict heart disease before a heart attack occurs accurately. This paper proposes an intelligent framework for heart disease prediction using edge computing, Cloud computing and ensemble learning techniques. The proposed system is evaluated with heart disease data and compared with traditional ensemble classifiers based on precision, weighting techniques and temporal metrics like arbitration delay and computational expense. The architecture also provides a facility for distributed learning at the node level, ensuring proper resource utilization and boosting accuracy, making it a suitable choice for health care and heavy-load applications. Accuracy of 96.5% was obtained based on the proposed intelligent framework for heart disease prediction at a reasonable latency, making this a unique pick compared to existing works.