Review Paper on Heart Attack Prediction System Using ML

Heart disease are a leading cause of death worldwide, making early detection critical for effectiveintervention. This project aims to develop a machine learning (ML) model to predict the likelihood of aheart attack based on various health and lifestyle parameters. The system features an interactive front-endapplication powered by Gradio, allowing users to input relevant health metrics and receive real-timepredictions. The ML model is trained using key algorithms such as Logistic Regression, Random Forest,and Gradient Boosting (XGBoost), validated on popular datasets like the UCI Heart Disease and Kaggle’sHeart Attack Analysis datasets. Additionally, advanced pre-trained generative models (LLMs) likeGemini are integrated to provide personalized lifestyle recommendations and explain predictions innatural language. The system outputs heart attack risk classifications, interpretable insights oncontributing factors, and actionable health advice to help users reduce their risk. By leveraging real-timefeedback, model explainability, and multimodal integration, this system aims to improve early detectionand empower users to make informed decisions about their heart health. The platform outputs riskclassifications as low, medium, or high, along with actionable insights into the most significantcontributing factors. By combining predictive accuracy, explainability, and personalized guidance, thesystem empowers users to make informed decisions about their heart health. It also providesrecommendations to adopt healthier lifestyle practices, such as dietary adjustments, increased physicalactivity, and stress management strategies. By harnessing the power of ML, real-time feedback, andnatural language processing, this system represents a significant step forward in the domain of predictivehealthcare. It has the potential to bridge the gap between early detection and proactive intervention,ultimately reducing the burden of heart disease on individuals and healthcare systems.