a-b-15/heart-attack-analysis

Heart Attack Risk Prediction

A machine learning application that predicts the risk of heart attack based on patient medical data.

Features

• —Interactive Web Interface: User-friendly form to input patient details
• —Real-time Predictions: Instant risk assessment with probability scores
• —High Accuracy: Trained on heart attack dataset with ~87% accuracy using SVM
• —Multiple Models Evaluated: Logistic Regression, Random Forest, SVM, and Neural Networks

Dataset

The model is trained on a comprehensive heart attack dataset with 305 patient records containing 14 features:

• —Age, Sex, Chest Pain Type
• —Resting Blood Pressure, Cholesterol
• —Fasting Blood Sugar, Resting ECG
• —Maximum Heart Rate, Exercise Induced Angina
• —ST Depression (Oldpeak), Slope
• —Number of Major Vessels, Thalassemia
• —Target (Heart Attack Risk)

Model Performance

• —Best Model: Support Vector Machine (SVM)
• —Accuracy: 86.89%
• —Precision: 87%
• —Recall: 87%

Usage

Simply input the patient's medical parameters in the web form and click "Predict Risk" to get an instant assessment.

Technology Stack

• —Backend: Flask (Python)
• —ML Libraries: scikit-learn, pandas, numpy
• —Frontend: HTML, Bootstrap 5
• —Deployment: Docker on Hugging Face Spaces