Skip to content

Riddhi147/MindLite

Repository files navigation

🧠 MindLite — Cognitive Health Monitoring System

MindLite is a comprehensive health-tech platform designed for the early detection and continuous monitoring of cognitive decline, specifically focusing on Alzheimer's disease. Using engaging, clinically-inspired games, MindLite tracks memory, attention, and cognitive performance over time.

🌟 Key Features

🎮 Cognitive Assessment Games

  • Family Recognition: Identify known faces and relationships (Visual Memory).
  • Pattern Recognition: Predict sequences and logical patterns (Logical Memory).
  • Memory Match: Speed-based symbol matching (Recall & Processing Speed).
  • Word Recall: Short-term and delayed verbal recall.
  • Reaction Time: Interactive tests for neuro-motor responses.

📊 Advanced Analytics

  • ML Predictions: Uses Scikit-learn models to generate cognitive risk scores.
  • Trend Analysis: Tracks performance over a minimum of 10 days to detect subtle declines.
  • Smart Alerts: Automatically notifies healthcare professionals and guardians if sudden drops in performance are detected.

🛠️ Tech Stack

Frontend

  • Framework: Next.js (React)
  • Styling: Tailwind CSS & Modern Glassmorphism UI
  • Components: Lucide Icons, Recharts for health telemetry

Backend (Architecture Refresh)

  • Next.js: Express.js server for robust data management.
  • Prisma ORM: Modern database access layer.
  • MySQL: Persistent relational storage for user profiles and health data.
  • Python: Specialized service for Machine Learning inference.

🚀 Getting Started

1. Prerequisites

  • Node.js (v18+)
  • MySQL
  • Python 3.9+ (for ML modules)

create DATABASE db_name; DATABASE_URL="mysql://root:password@localhost:3306/db_name"

3. Installation

Frontend: npm install

Backend: npm install npx prisma generate

4. Running the Application

uvicorn main:app --reload npm run dev

About

MindLite is a web app designed to detect early signs of Alzheimer's disease by tracking cognitive performance through simple games based on the Montreal Cognitive Assessment (MoCA). Scores are converted into a Z score and analyzed with a random forest model over time to detect sudden or gradual decline and alert doctors or guardians.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors