Optimized CNN Framework for Heart Attack Detection in Consumer Healthcare Systems Using Flower Pollination Algorithm

One of the primary causes of death globally, heart disease still affects millions of people; thus, early identification is very essential for efficient treatment. To meet the need for precise, real-time predictions, this study proposes an improved CNN framework for heart attack diagnosis in Consumer Healthcare Systems (CHS). Using a Kaggle comprehensive dataset including age, cholesterol levels, and electrocardiographic findings, we employed the Flower Pollination Algorithm (FPA) to maximize important hyperparameters like learning rate and dropout rate. Our suggested approach far exceeded conventional machine learning and deep learning models like GRU, LSTM, and logistic regression. These results show the possibility of FPA-optimized CNNs in improving CHS heart attack prediction.