AI-Powered Liver Cirrhosis Risk Prediction

A clinical decision-support tool that predicts liver cirrhosis using key blood indicators and lifestyle factors. Built with Flask, trained with a Random Forest classifier.

Project Lead: Katireddy Rajeswari | SmartBridge Internship, IIC 2025

Key Features

Simple Web Form

Users can input patient data through a clean, mobile-friendly interface.

High Accuracy

Trained on clinical features with 100% accuracy on training data (with validation).

Fast Prediction

Real-time prediction powered by a Random Forest model and Scikit-learn.

7 Medical Features

Based on expert-chosen features: Age, Albumin, Bilirubin, SGPT, etc.

Dynamic Feature Handling

App loads model-required fields automatically using feature_names.json.

Deploy-Ready

Includes requirements.txt, Procfile, and README for one-click deployment.

Technology Stack

Python Flask Scikit-learn Pandas Joblib HTML + Bootstrap Render (Hosting)

How to Run

1. Install dependencies:

pip install -r requirements.txt

2. Start the Flask app:

python app.py

3. Open your browser:

http://127.0.0.1:5000

Future Vision

📄 Export Results: Add PDF or CSV report generation.

📊 Visual Feedback: Add charts for model confidence levels.

🧠 Model Improvements: Include explainable AI for transparency.

🌐 Public Deployment: Deploy app on Render or Streamlit Cloud.