Transformative AI Solutions: A Hybrid Diagnostic Model for Cardiovascular Disease Utilizing Comprehensive Feature Analysis

Abstract

CVD is the number one disease that causes death around the globe, making it crucial to develop early and precise prediction models which can help in prevention. In this paper, we propose a hybrid deep learning model with CNN and Multilayer Perceptron (MLP) Neural Networks for CVD classification. The model aims to improve the predictive performance of CVD by using important patient health metrics like blood pressure, cholesterol level, BMI, and lifestyle choices. Utilizing CNN deep learning techniques on Processed Medical Data allows us to extract sophisticated features and patterns. Simultaneously, MLP integrates complex feature interaction which further enhances classification performance. Proposed system is evaluated with accuracy, precision, recall, f1 score and area under curve of Receiver Operating Characteristics (AUC-ROC) to achieve results in standalone models. Results showed that the hybrid CNN- MLP model outperforms the fundamental models and achieves the set benchmarks. These results imply the utility of the CNN- MLP hybrid model as a powerful, easy-to-use, and efficient CVD decision-support system for early AI diagnosis. This work assists in personalized medicine, telehealth and urgent healthcare service delivery to advance the use of AI technologies for automation of medical diagnostics.