An Intelligent IoT-Enabled Healthcare Framework for Early Cardiovascular Disease Detection Using a Hybrid Deep Learning Model

Cardiovascular disease (CVD) is among the globe's major causes of death, and early diagnosis is critical in the battle against mortality and better patient outcomes. Through the convergence of healthcare services with Internet of Things (IoT) technology, the intersection of real-time monitoring and sophisticated data analysis methods in CVD diagnosis is paramount in successful management of patients. This work presents a new method of CVD detection using a hybrid deep model based on a Vision Transformer (ViT) and an LSTM network. The suggested system utilizes spatial feature extraction using ViT and temporal dependency management using LSTM in a manner to produce a system that is extremely effective in dealing with complex and heterogeneous medical data. The given model is intended to handle several parameters of health such as age, blood pressure, cholesterol, heart rate, sugar level, smoker or not, exercise habit, family history, and BMI. By including these parameters, the model is able to make precise and timely predictions for CVD and can have applications in being used in IoT-based healthcare systems offering real-time monitoring and diagnosis. Through proper experimentation and comparison with existing CVD detection techniques, the proposed ViT-LSTM model performs better than conventional methods in important performance measures such as detection accuracy, time complexity, error rate, precision, and recall.