IOT-DRIVEN HEART DISEASE PREDICTION WITH INTELLIGENT CLASSIFIER AND SQUIRREL SEARCH FEATURE SELECTION

Abstract

Cardiovascular disease (CVD) is the leading cause of global mortality in the modern world. This situation is difficult to predict and requires a combination of advanced techniques and specialist knowledge. Healthcare systems have recently adopted the Internet of Things (IoT) to collect critical sensor data to diagnose and predict CVD. Predictive models can be made more accurate and effective through such integration, which could radically change how we manage cardiovascular health. This study presents an improved squirrel search optimization algorithm for searching vital indications of CVD. To address the issue of low-cardiac diagnostic accuracy, the proposed IoT system uses enhanced squirrel search optimization with deep convolutional neural networks (SSO-DCNN). This new approach uses data from smartwatches and cardiac devices, which monitor patients’ electrocardiogram (ECG) and blood pressure readings. The proposed SSO-DCNN performs well compared to well-known deep learning networks such as logistic regression. The findings show an accuracy of 99.1% over current classifiers, suggesting effectiveness in the CVD prediction.