How techstack-ujjwal/heart_disease_prediction_system works
This repository’s visible product is a Streamlit heart-disease risk prediction app. A user enters patient details, the app encodes them into model columns, scales numeric features, and runs a prediction; it then displays a risk category and probability with a medical disclaimer. The checked-in model and preprocessing artifacts are loaded by the app. The README provides no further capability details, and the notebook and dataset were not sampled, so their roles and connections are omitted.
Main entry points: app.py Feature Encoding · scaler.pkl Feature Scaler · heart_disease_model.pkl Trained Model · columns.pkl Expected Columns
Components 9
User Interface
- Patient Input Form
app.py - Risk Result
app.py
Prediction Processing
- Feature Encoding
app.py - Numeric Scaling
app.py - Risk Inference
app.py
Model Artifacts
- Feature Scaler
scaler.pkl - Trained Model
heart_disease_model.pkl - Expected Columns
columns.pkl
Other components
- User
Connections 8
- User → Patient Input Form: enters details
app.py - Patient Input Form → Feature Encoding: submits details
app.py - Feature Encoding → Expected Columns: aligns columns
app.py - Feature Encoding → Numeric Scaling: passes features
app.py - Numeric Scaling → Feature Scaler: transforms with
app.py - Numeric Scaling → Risk Inference: passes scaled data
app.py - Risk Inference → Trained Model: predicts with
app.py - Risk Inference → Risk Result: displays prediction
app.py