A Smart Healthcare Application for Early Heart Disease Prediction: Leveraging Feature Optimization and Deep Learning Techniques

Internet of Things (IoT) and Information Technology (IT) are integrated and released as many browser-based applications that can fulfil the needs of people in this smart world. Especially, these technologies are used in medical field widely to provide the rich medical facilities and services to the society. Even though, the society is not fulfilled in the medical facility due to the lack of awareness and fast services. So that the people are trailing their living days every day. This is the right time to enrich the medical facilities by providing rich facility and services through available technologies. This paper suggests a novel smart healthcare system that employs feature optimization and an ensemble deep classifier to forecast illnesses, including heart disease, cancer, and diabetes, as a result of its perseverance. The optimization technique Random Search Algorithm (RSA) was used in this study to identify the most impactful features for improving the prediction accuracy. A new ensemble deep classifier was also developed to effectively diagnose diseases. The classifier was created by combining the RSA base Fuzzy Decision Tree Algorithm (RFDTA) with the Convolutional Neural Network (CNN). The system’s performance was evaluated through various experiments and found to be both efficient and effective, outperforming existing healthcare systems.