README.md

metadata

title: AIO2025M03 Heart Disease Diagnosis
emoji: 🫀
colorFrom: blue
colorTo: red
sdk: gradio
sdk_version: 5.38.2
app_file: app.py
pinned: false

🫀 Cleveland Heart Disease Diagnosis (Ensemble Demo)

An interactive demo that predicts heart disease risk from patient features using an ensemble of ML models trained on the Cleveland dataset.

✨ Features

• 🧩 Ensemble models: Decision Tree, k-NN, and Naive Bayes with soft voting
• 📋 Validation metrics: ROC-AUC, Accuracy, Precision, Recall, F1 (80/20 stratified split)
• 🧪 Patient inputs: 13 clinical features with sensible defaults and example selector
• 📈 Results visualization: Color/emoji-coded prediction confidence bar chart per model
• ℹ️ Clear context: Styled About card and fixed disclaimer (educational use only)
• 🎨 Theming: VLAI template with configurable colors, fonts (Inter), header/footer

🚀 Quick Start

Local Installation

git clone <repository-url>
cd AIO2025M03_HEART_DISEASE_PREDICTION
pip install -r requirements.txt
python app.py

Usage

1. Ensure data/cleveland.csv exists (included in this repo)
2. Launch the app: python app.py
3. Review the About card and disclaimer at the top
4. Inspect data preview and validation metrics (trained at startup, 80/20)
5. Enter patient features or pick an example, then click Predict
6. Read individual model results and the predictions overview chart

🧠 Technical Highlights

• Preprocessing: Numeric median imputation; categorical most-frequent + one-hot
• Pipelines: Shared preprocessor wrapped with each estimator; ensemble via VotingClassifier
• Evaluation: Stratified train/validation split, with ROC-AUC and common classification metrics
• Visualization: Plotly bar chart; Gradio Blocks UI using VLAI template

📋 Requirements

• Python 3.8+
• Gradio 5.38+
• Scikit-learn
• Pandas
• NumPy
• Plotly

🎨 Theming & Branding

The UI is powered by vlai_template.py and can be customized programmatically:

import vlai_template as vt

vt.configure(
    project_name="Heart Disease Diagnosis Project",
    year="2025",
    module="03",
    description=(
        "Predict heart disease risk from patient data with ML models "
        "trained on the Cleveland dataset."
    ),
    colors={
        "primary": "#0F6CBD",
        "accent": "#C4314B",
        "bg1": "#F0F7FF",
        "bg2": "#E8F0FA",
        "bg3": "#DDE7F8",
    },
    font_family=(
        "'Inter', -apple-system, BlinkMacSystemFont, 'Segoe UI', Roboto, "
        "'Helvetica Neue', Arial, 'Noto Sans', 'Liberation Sans', sans-serif"
    ),
    meta_items=[
        ("Dataset", "Cleveland Heart Disease"),
        ("Models", "Decision Tree, k-NN, Naive Bayes"),
        ("Ensemble", "Soft Voting"),
    ],
)

You can also inject the info and disclaimer blocks:

gr.HTML(vt.render_info_card(icon="🫀", title="About this demo"))
gr.HTML(vt.render_disclaimer(
    text=(
        "This interactive heart disease prediction demo is provided strictly "
        "for educational purposes and is not intended for clinical use."
    )
))

📄 License

Educational use for AIO2025 course materials. This demo is not medical advice.

---

Live Demo: coming soon