❤️ CardioCare: Predicting Heart Disease Risk with Machine Learning & FastAPI

📌 Short Description
CardioCare develops a predictive machine learning model to identify individuals at risk of heart disease using clinical and behavioral health indicators. It demonstrates a complete end-to-end workflow—from exploratory data analysis and feature engineering to model training, evaluation, and API deployment with FastAPI.

---

📊 Project Overview

This project highlights data science, machine learning, and deployment skills through a healthcare-focused classification problem. Beyond notebooks, the trained model is wrapped in a FastAPI application with Pydantic validation and deployed as a REST API, making it usable in real-world scenarios.
👉 You can go through the project files for detailed instructions on usage and testing.

---

🚀 Key Steps & Skills Demonstrated

📂 Data Loading & Exploration

• Loaded structured health data with pandas.
• Performed descriptive analysis (.info(), .describe()) to understand data distribution.
• Summarized missing values and identified patterns.

🧹 Data Cleaning & Preparation

• Removed records with missing values.
• Standardized features for training.

📊 Exploratory Data Analysis (EDA)

• Visualized relationships between risk factors and heart disease (gender, age groups, BMI, glucose, smoking, diabetes).
• Tools: matplotlib, seaborn.

🔍 Feature Selection

• Applied correlation analysis to identify top predictors.

⚖️ Handling Imbalanced Data

• Applied SMOTE to balance dataset classes and improve fairness.

🧠 Model Training

• Trained and compared two models:
  • 🌲 Random Forest Classifier
  • 📍 K-Nearest Neighbors (KNN)

📈 Model Evaluation

• Evaluated models using:
  • Accuracy Score
  • Confusion Matrix
  • Precision, Recall, F1-Score

🌐 Model Deployment

• Saved trained model using pickle.
• Built a FastAPI app with:
  • /predict/ endpoint for predictions
  • Pydantic validation for clean input handling
• Tested the API with Python requests.

---

📌 Skills Demonstrated

• Data Wrangling & Analysis: pandas, numpy
• Statistical Visualization: matplotlib, seaborn
• Exploratory Data Analysis (EDA) & Feature Engineering
• Class Imbalance Handling: SMOTE
• Machine Learning Classification: Random Forest, KNN (scikit-learn)
• Model Evaluation: Accuracy, Precision, Recall, F1-Score
• Deployment & API Development: FastAPI, Pydantic, Pickle, Docker
• API Testing: requests

---

🧰 Tools & Technologies Used

• Python 3.11
• Libraries: pandas, numpy, matplotlib, seaborn, scikit-learn, imblearn, FastAPI, Pydantic, requests

---

📂 File Structure

• EDA_and_HeartDisease_prediction.ipynb → Jupyter Notebook containing EDA, feature engineering, model training, and evaluation
• model_api.py → FastAPI application for serving the trained ML model as a REST API
• testing.py → Client script to test API endpoints with sample input data
• Dockerfile → Docker configuration file to containerize the FastAPI application
• model.pkl → Serialized trained machine learning model
• dataset.csv → Raw dataset used for training and analysis
• • Other supporting files → Pickled trained model, dataset.

---

🔗 Note: Please check the repository files for detailed instructions on how to run the project and test the API.