Stroke _prediction_ML

AI Stroke Prediction System is a machine learning-based web application designed to assess stroke risk based on patient demographic and health data. The system addresses extreme class imbalance using advanced resampling techniques to ensure high accuracy and reliability in medical risk detection. Key Features: Accurate Risk Assessment: Utilizes Machine Learning classification models to accurately predict the likelihood of a stroke based on patient health indicators. Class Imbalance Handling: Applied SMOTE (Synthetic Minority Over-sampling Technique) to resolve severe data imbalance, improving sensitivity and reducing false negatives for stroke cases. Fast & Scalable API: Built a high-performance RESTful API using FastAPI to serve real-time predictions with low latency. End-to-End Pipeline: Complete Machine Learning workflow including Exploratory Data Analysis (EDA), data cleaning, feature engineering, model training, evaluation, and API deployment. Tech Stack & Tools: Programming Language: Python API Framework: FastAPI Machine Learning: Scikit-Learn, SMOTE (Imbalanced-Learn) Data Processing & Analysis: Pandas, NumPy, Seaborn, Matplotlib