CARDIO-CARE: A Deep Learning-Powered IoT Recommender System for Cardiovascular Disease Classification

cardiovascular disease (CVD) is a major global health issue, highlighting the necessity of accurate early detection and personalized treatment plans. This work introduces CARDIO-CARE, an advanced framework for CVD prediction, along with customized recommendations to improve patient care and diagnostic accuracy. The proposed methodology used an attention-integrated Long Short-Term Memory (LSTM) network to enhance CVD prediction and accurately classify six conditions such as atrial fibrillation (AF), acute coronary syndrome (ACS), hypertension (HTN), left ventricular hypertrophy (LVH), ischemic heart disease (IHD), and myocardial infarction (MI). In addition, based on classified output class, this framework includes a recommender system that offers lifestyle and dietary guidelines. Using the UCI cardiac disease dataset, the experimental findings confirmed the model’s efficiency with a classification accuracy rate of 99% and a precision of 100%. The result showed that the suggested approach surpassed existing methodologies, making it a valuable tool for patient-specific healthcare recommendations and early CVD detection.