INTELLICARDIA: AN ARTIFICIAL INTELLIGENCE BASED DECISION SUPPORT SYSTEM FOR PREDICTING HEART DISEASE RISK

Abstract

Cardiovascular diseases (CVDs) are complex, multifactorial conditions that include disorders of heart and blood vessels. According to the statistics report of World Health Organization almost billions of people die every year due to CVD. In order to control the mortality rate there is need for early prediction of disease before it becomes a life-threatening issue. Therefore, there is a need for an Intelligent automated system that can assist in early screening and risk assessment. To overcome the limitations of previous models, this research presents an advanced clinical decision support system (CDSS) designed for the early prediction and risk stratification of heart disease using a hybrid Artificial Intelligence (AI) framework. The system received training and evaluation through a merged dataset which combined 5,158 patient records from the Cleveland Heart Disease dataset and the Framingham Heart Study using clinical parameters such as age, sex, systolic blood pressure, cholesterol level, heart rate, and diabetes status. An expert driven fuzzy inference system with data-driven machine learning engine is utilized to improve computational robustness, The Preprocessing techniques, Min-Max, normalization and (SMOTE) Synthetic minority over sampling technique is used to resolve class imbalance in the medical data. This system is evolved through three versions and this is 3.0 version that is based on soft voting machine learning technique consisting of XGboost, Random Forest, Support Vector Machine and Logistic Regression. please the weighted fusion mechanism is utilized that provide statistical ML probabilities with linguistic fuzzy reasoning that classifies the patient into Low, Moderate and high risk class. The system bridges the gap between the computational output and clinician trust with explainable feature using explainable AI powered by XAI and large language model Google Gemini 3.2, which translates features into human readable rationale, the system is deployed via streamlit and provide real time visual analytics, patient records archiving and automated PDF diagnostic report. This hybrid framework achieves a peak predictive accuracy of 94 to 95%. IntelliCardia is a transparent, scalable, and a reliable solution that can be used for predicting the cardiovascular disease to prevent the mortality rate. By utilizing this CDSS, a person can know their heart risk level and by following the advices given in PDF report they can minimize their risk of Cardiac arrest and Heart failure.