Cardiovascular risk detection using Harris Hawks optimization with ensemble learning model on PPG signals

Cardiovascular (CVD) risk detection using Electrocardiography (ECG) and photoplethysmography (PPG) signals is an emerging field of research in the area of machine learning and biomedical engineering. ECG is an electrical measurement that captures cardiac actions and is the gold standard for identifying CVD. But, ECG cannot be used for continuous cardiac monitoring because of its necessity for user participation. PPG is an optically attained signal for detecting blood volume changes in the microvascular bed of tissues. Deep learning (DL) methods have shown remarkable performance in predicting and detecting CVD disease using PPG signals. Therefore, this study designs a new Cardiovascular Risk Detection using Harris Hawks Optimization with Ensemble Learning (CRDHHO-EL) model on PPG Signals. The presented CRDHHO-EL technique examines the PPG signals to identify the risks of cardiovascular diseases. To accomplish this, the CRDHHO-EL technique uses an ensemble of three classifiers, namely deep belief network (DBN), deep auto-encoder (DAE), and extreme learning machine (ELM) models. Moreover, the HHO algorithm is used for adjusting the hyperparameter values of the classifier models, boosting overall performance. The experimental result analysis of the CRDHHO-EL on the open-access dataset showcases the significant performance of the CRDHHO-EL technique over other existing approaches.