Skip to content

Latest commit

 

History

38 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

🧠 CuraMind

AI-Assisted Early Cancer Risk Screening & Clinical Decision Support

Live Demo Backend API

CuraMind is an AI-assisted preventive healthcare platform designed to simplify the early cancer screening journey.

It collects relevant patient information—including demographics, lifestyle factors, family history, medical information, environmental factors, and symptoms—and processes the assessment through CuraCore™, our rule-based risk screening engine.

The system generates cancer-wise risk scores, explains contributing factors, and provides personalized screening-oriented recommendations.

⚕️ Important: CuraMind is a screening and clinical decision-support prototype. It does not diagnose or predict cancer and does not replace qualified medical professionals or clinical evaluation.


🚀 Live Prototype

🌐 Frontend

https://curamind-ai.netlify.app/

⚙️ Backend API

https://curamind-ai-mx76.onrender.com/

❤️ Health Check

https://curamind-ai-mx76.onrender.com/health

The deployed prototype connects the React frontend to the FastAPI backend and runs the CuraCore risk-screening workflow end-to-end.


🎯 Problem

Cancer screening can be confusing and delayed because people may:

  • Ignore early symptoms
  • Be unaware of important risk factors
  • Not know which screening pathway is appropriate
  • Search for information across disconnected sources
  • Reach healthcare providers later than ideal

Healthcare providers can also benefit from better-organized patient information and risk-screening support.

CuraMind aims to simplify this journey by bringing patient assessment, risk screening, explanations, and recommendations into one workflow.


💡 Solution

CuraMind follows a patient-to-screening workflow:

👤 Patient
    ↓
📋 Health Assessment
    ↓
🧠 CuraCore™ Risk Engine
    ↓
🎯 Cancer-Wise Risk Results
    ↓
💡 Personalized Recommendations
    ↓
👨‍⚕️ Clinical Review

The current prototype focuses on the first five stages and provides a working end-to-end screening workflow.


🧠 CuraCore™

CuraCore™ is the current intelligence layer of CuraMind.

The present prototype uses a transparent rule-based risk engine rather than a trained machine-learning model.

Current pipeline

Patient Data
     ↓
Configured Risk Rules & Weights
     ↓
Cancer-Wise Risk Scores
     ↓
Risk Classification
     ↓
Explanation Engine
     ↓
Recommendation Engine
     ↓
Final Screening Report

Risk Engine

The risk engine evaluates configured rules against patient information and symptom data to calculate cancer-wise scores.

Explanation Engine

The explanation engine identifies the matched risk factors and generates understandable reasons associated with the calculated scores.

Recommendation Engine

The recommendation engine uses the resulting risk level and configured cancer-specific information to generate:

  • Recommended actions
  • Next steps
  • Relevant specialists
  • Lifestyle guidance
  • Tests to discuss where applicable

✨ Current Features

👤 Patient Assessment

The assessment workflow collects structured information across multiple areas:

  • Personal information
  • Medical information
  • Lifestyle
  • Family history
  • Environmental factors
  • Symptoms
  • Relevant reports/information

🧠 Rule-Based Risk Screening

CuraCore™ evaluates the submitted information using configured rules and risk weights.

🎯 Cancer-Wise Risk Assessment

The backend calculates scores for the supported cancer categories and classifies applicable results into risk levels.

🔍 Explainable Results

The system provides the factors/rules that contributed to the calculated risk score.

💡 Personalized Recommendations

Recommendations are generated according to the resulting risk level and cancer-specific configuration.

📊 Results Dashboard

The frontend presents:

  • Overall risk
  • Cancer-wise scores
  • Risk levels
  • Contributing factors
  • Recommended actions
  • Suggested tests
  • Lifestyle guidance

🔄 Full-Stack Integration

The frontend communicates with the FastAPI backend through REST API endpoints.


🖥️ How CuraMind Works

🏠 1. Start Screening

Begin the CuraMind cancer-risk screening journey by clicking "Start Screening" on the landing page.

CuraMind Landing Page

📋 2. Health Assessment

Complete a guided 7-step assessment covering personal, lifestyle, family, medical, symptom, and environmental factors.

CuraMind Health Assessment

📄 3. Medical Reports

Optionally upload supporting medical reports before clicking "Run CuraCore™ Analysis" to begin the screening process.

CuraMind Medical Reports

🧠 4. CuraCore™ Analysis

The rule-based CuraCore™ engine evaluates submitted risk factors and generates screening results.

CuraCore Analysis

🎯 5. Risk Dashboard

View overall risk, cancer-wise scores, priority areas, recommended specialists, and contributing risk factors.

CuraMind Risk Dashboard

📑 6. Final Report

Review, share, print, or download the generated CuraCore™ screening report.

CuraMind Final Report CuraMind Cancer Wise Report

🏗️ System Architecture

Final CuraMind prediction flow

User completes CuraMind assessment
             ↓
Frontend sends all assessment data to backend
             ↓
Shared preprocessing layer
  • age
  • sex
  • BMI
  • smoking
  • lifestyle
  • medical history
             ↓
CuraMind model router
             ↓
  ┌──────────────────────────────┐
  │ Lung Cancer Model       → Risk score │
  │ Breast Cancer Model      → Risk score │
  │ Oral Cancer Model        → Risk score │
  │ Colorectal Cancer Model  → Risk score │
  │ Cervical Cancer Model    → Risk score │
  │ Prostate Cancer Model    → Risk score │
  └──────────────────────────────┘
             ↓
Risk validation + normalization
             ↓
Risk factor breakdown dashboard

Example prediction output

If a user submits their assessment, the frontend can display the model results as a cancer-wise breakdown:

Lung Cancer        → 72
Breast Cancer      → 14
Oral Cancer        → 8
Colorectal Cancer  → 31
Cervical Cancer    → N/A
Prostate Cancer    → N/A

The dashboard presents these scores visually alongside the contributing risk factors, explanations, and recommendations.


🛠️ Technology Stack

Frontend

  • React
  • TypeScript
  • Vite
  • Tailwind CSS
  • React Router
  • Framer Motion
  • Lucide React
  • Three.js / Spline tooling

Backend

  • Python
  • FastAPI
  • Uvicorn
  • Pydantic

Risk & Intelligence Layer

  • CuraCore™ Rule-Based Risk Engine
  • Risk Weight Configuration
  • Explanation Engine
  • Recommendation Engine
  • JSON-based configuration

Deployment

  • Netlify — Frontend
  • Render — Backend

📁 Project Structure

CuraMind/
│
├── backend/
│   ├── api/
│   │   └── routes.py
│   │
│   ├── config/
│   │   ├── patient_schema.json
│   │   ├── recommendations.json
│   │   └── risk_weights.json
│   │
│   ├── exceptions/
│   │   └── handlers.py
│   │
│   ├── models/
│   │   ├── request_models.py
│   │   └── response_models.py
│   │
│   ├── services/
│   │   ├── risk_engine.py
│   │   ├── explanation_engine.py
│   │   └── recommendation_engine.py
│   │
│   ├── utils/
│   │   └── responses.py
│   │
│   ├── main.py
│   ├── requirements.txt
│   └── test_engine.py
│
├── frontend/
│   ├── public/
│   ├── src/
│   │   ├── api/
│   │   ├── assets/
│   │   ├── components/
│   │   ├── context/
│   │   ├── hooks/
│   │   ├── lib/
│   │   ├── pages/
│   │   ├── types/
│   │   ├── App.tsx
│   │   └── main.tsx
│   │
│   ├── package.json
│   ├── package-lock.json
│   ├── vite.config.ts
│   └── tsconfig.json
│
├── patient_schema.md
└── README.md

🔌 API

The backend currently exposes the following primary endpoints:

Method Endpoint Purpose
GET / API information
GET /health Health check
POST /predict Process patient assessment and generate screening results

Example health response

{
  "success": true,
  "message": "API is healthy.",
  "data": {
    "status": "healthy"
  }
}

💻 Local Development

Follow the steps below in order to run CuraMind locally.


1. Prerequisites

Before starting, make sure your computer has the following installed:

  • Python 3.10 or newer
  • Node.js 18 or newer
  • Git
  • VS Code (recommended)

Check if they are installed

Open a terminal in VS Code and run:

python --version
node --version
npm --version
git --version

If any command says that it is not recognized, install the missing software first.

Windows

If winget is available, you can install the required tools with:

winget install Python.Python.3.10
winget install OpenJS.NodeJS.LTS
winget install Git.Git

After installation, restart VS Code and check the versions again.

macOS

If you use Homebrew:

brew install python
brew install node
brew install git

Then verify:

python3 --version
node --version
npm --version
git --version

2. Clone the repository

Clone the CuraMind repository and open the project folder:

git clone https://github.com/javedarham2-coder/CuraMind-AI.git
cd CuraMind-AI

Make sure you are now inside the main CuraMind-AI folder.


3. Create the Python virtual environment

⚠️ Important: Create the virtual environment inside the main CuraMind-AI folder, NOT inside backend.

Run this while you are in the project root:

python -m venv .venv

After this, your project should look like:

CuraMind-AI/
├── .venv/
├── backend/
├── frontend/
└── README.md

4. Activate the virtual environment

Windows PowerShell

Run:

.\.venv\Scripts\Activate.ps1

If PowerShell blocks the activation script, run:

Set-ExecutionPolicy -Scope Process -ExecutionPolicy Bypass

Then activate again:

.\.venv\Scripts\Activate.ps1

You should see (.venv) at the beginning of your terminal.

macOS / Linux

Run:

source .venv/bin/activate

5. Start the backend

Now enter the backend folder:

cd backend

Install the backend dependencies:

pip install -r requirements.txt

Start the FastAPI backend:

uvicorn main:app --reload

The backend will normally run at:

http://127.0.0.1:8000

Keep this terminal running.


6. Start the frontend

Open a new terminal in VS Code.

Go back to the project root:

cd CuraMind-AI

Then enter the frontend folder:

cd frontend

Install the frontend dependencies:

npm install

7. Configure the frontend API URL

Before starting the frontend, you must create a .env file inside the frontend folder. This tells CuraMind where the local backend is running.

First, make sure you are inside the frontend folder:

cd frontend

Create a new file named exactly:

.env

Your folder should look like this:

CuraMind-AI/
├── backend/
└── frontend/
    ├── .env
    ├── package.json
    └── ...

Open the .env file and paste exactly this line:

VITE_API_URL=http://127.0.0.1:8000

Save the file.

⚠️ Important: The file must be named .env — not .env.txt, and it must be inside the frontend folder.

This connects the frontend to the local FastAPI backend.


8. Start the frontend

Run:

npm run dev

Vite will show a local URL in the terminal, usually:

http://localhost:5173

Open that URL in your browser.


9. Local development complete 🎉

You should now have:

Frontend → http://localhost:5173
              ↓
Backend  → http://127.0.0.1:8000

Keep both terminals running while using CuraMind locally.

If you make code changes

  • Frontend changes are automatically refreshed by Vite.
  • Backend changes are automatically reloaded by Uvicorn.
  • If you pull new changes from GitHub, run:
git pull

Then install any new dependencies if requirements.txt or package.json has changed:

pip install -r backend/requirements.txt

and/or:

cd frontend
npm install

🌐 Production Deployment

The current prototype uses:

GitHub
   │
   ├── Frontend → Netlify
   │
   └── Backend  → Render

The frontend communicates with the deployed backend through:

VITE_API_URL=https://curamind-ai-mx76.onrender.com

Deployment workflow

Edit Code
    ↓
Test Locally
    ↓
Git Commit
    ↓
Git Push
    ↓
GitHub
    ├──────────────→ Netlify
    │                 Frontend Deployment
    │
    └──────────────→ Render
                      Backend Deployment

🧪 Current Prototype Status

🟢 Working

  • Patient assessment
  • Frontend/backend integration
  • FastAPI API
  • CuraCore rule-based risk engine
  • Cancer-wise risk scoring
  • Risk classification
  • Explanation generation
  • Personalized recommendations
  • Results/report interface
  • End-to-end deployed workflow

🟡 Not currently implemented

  • Doctor/clinical dashboard
  • Persistent patient database
  • Production authentication
  • Advanced hospital/doctor locator functionality
  • Full production-grade healthcare infrastructure

🔵 Planned Future Enhancements

  • Curated cancer-risk datasets
  • Unified machine-learning training dataset
  • Machine-learning-based risk estimation
  • Random Forest / Gradient Boosting model comparison
  • SHAP-based explainability
  • Medical report analysis
  • Expanded clinical workflows
  • Production-grade security and deployment

The current hackathon prototype intentionally uses a rule-based engine so that the screening logic remains transparent and explainable.


🛣️ Roadmap

Phase 1 — Assessment

Patient registration and structured health assessment.

Phase 2 — CuraCore™ Screening

Rule-based cancer risk screening and risk stratification.

Phase 3 — Personalized Guidance

Screening-oriented recommendations and explainable results.

Phase 4 — Clinical Workflow

Doctor dashboard, patient prioritization, clinical validation, and follow-up planning.

Phase 5 — Future AI Enhancement

Machine-learning models, SHAP explainability, medical report analysis, and broader deployment.


⚕️ Medical Disclaimer

CuraMind is an AI-assisted screening and clinical decision-support prototype.

It does not diagnose cancer, confirm the presence of cancer, or replace professional medical evaluation.

The risk scores and recommendations generated by CuraCore™ are intended to support screening awareness and decision-making. Users should consult qualified healthcare professionals for appropriate medical assessment, testing, diagnosis, and treatment.


Made with ❤️ for more accessible and explainable preventive healthcare.

CuraMind — Earlier Screening. Better Decisions. Better Outcomes.

Intellectual Property

© 2026 CuraMind Team. All Rights Reserved.

CuraMind and CuraCore™ are proprietary project names and materials developed by the CuraMind Team.

This repository is publicly available for hackathon evaluation and demonstration. No license is granted for reuse, redistribution, commercialization, or derivative works except where explicitly stated in the repository.

👥 Team Details

Team Name

The 3 Bytes

Project Name

CuraMind — AI-Assisted Early Cancer Risk Screening & Clinical Decision Support

Track

HealthTech & Social Impact

Team Lead

Mohammad Arham Javed
Role: Team Lead & Full-Stack Developer
Contact: +91 7905833216

Team Members

Name Role Contact
Mohammad Raiyan Frontend Developer +91 9336078040
Avnee Shukla Presentation & UI/UX Designer +91 9651440786

About

Prototype Project Development

Resources

Stars

2 stars

Watchers

1 watching

Forks

Releases

Packages

Contributors

Languages