An industry-grade, production-ready AI healthcare platform capable of predicting multiple diseases using patient medical data and advanced machine learning models. Built for hospitals, diagnostic labs, and healthcare startups.
- Multi-Disease Support: Predicts Heart Disease, Diabetes, Breast Cancer, Kidney Disease, and Liver Disease.
- Explainable AI (XAI): Integrated SHAP (SHapley Additive exPlanations) to explain model predictions visually.
- Premium UI/UX: A futuristic glassmorphism dashboard built with Streamlit and custom CSS.
- Automated Clinical Reports: Generates downloadable PDF reports with patient info, prediction risks, and health recommendations.
- Advanced EDA Dashboard: Interactive Plotly visualizations for dataset insights, feature correlations, and target distributions.
- Robust ML Pipeline: Automated preprocessing, scaling, and training using Random Forest, XGBoost, and Logistic Regression to select the best performing model.
AI_MultiDisease_Healthcare_Platform/
│
├── data/ # Contains raw CSV datasets
├── models/ # Serialized best-performing ML models (.pkl)
├── reports/ # Generated PDF clinical reports
├── components/ # UI components
├── styles/
│ └── style.css # Premium glassmorphism styling
├── utils/
│ ├── data_loader.py # Automated dataset downloading from OpenML/sklearn
│ ├── preprocess.py # Data cleaning and scikit-learn preprocessing pipelines
│ ├── predict.py # Inference and caching logic
│ ├── explainability.py # SHAP integration for model explainability
│ └── report_generator.py # FPDF logic for PDF report generation
│
├── app.py # Main Streamlit application
├── config.py # Global configurations
├── train_models.py # Script to train and evaluate all ML models
├── requirements.txt # Project dependencies
└── README.md # Documentation
- Clone the repository:
git clone https://github.com/yourusername/AI_MultiDisease_Healthcare_Platform.git
cd AI_MultiDisease_Healthcare_Platform- Create a virtual environment:
python -m venv venv
source venv/bin/activate # On Windows: .\venv\Scripts\activate- Install dependencies:
pip install -r requirements.txt- Download datasets:
python utils/data_loader.py- Train the models:
python train_models.py- Run the application:
streamlit run app.py- Heart Disease: UCI Heart Disease Dataset (OpenML)
- Diabetes: Pima Indians Diabetes Database (OpenML)
- Breast Cancer: Wisconsin Breast Cancer Dataset (scikit-learn)
- Kidney Disease: Chronic Kidney Disease Dataset (OpenML)
- Liver Disease: Indian Liver Patient Dataset (ILPD) (OpenML)
This project is fully ready to be deployed on Streamlit Community Cloud, Render, or HuggingFace Spaces.
Simply connect your GitHub repository to Streamlit Cloud and set app.py as the entry point.
[Your Name / Your GitHub Profile] Machine Learning Engineer | Data Scientist | Full Stack AI Developer
Developed as part of the CodeAlpha Machine Learning Internship