AI Health Partner

Health-Shield is a multi-platform health assistant application with a backend API, web frontend, and mobile app. It leverages AI and machine learning to provide health-related services, including chat, symptom assessment, and predictions.

Project Structure

Health-Shield/
├── backend/      # Python Flask API and ML models
├── frontend/     # React web app (Vite)
├── mobile/       # React Native mobile app (Expo)

Backend

• Location: backend/
• Tech: Python, Flask, SQLAlchemy, OpenAI, Google GenAI, ML libraries
• Features:
  • REST API for health chat, authentication, prediction, transcription
  • ML models for symptom analysis and health prediction
  • Database integration (SQLite)
• Key files:
  • main.py: Main Flask app
  • requirements.txt: Python dependencies
  • app/: Core modules (routes, models, database)
  • Dockerfile: Containerization

Backend Setup

1. Navigate to backend/
2. Create and activate virtual environment:

   python3 -m venv env
   source env/bin/activate

3. Install dependencies:

   pip install -r requirements.txt

4. Run the server:

   python main.py

5. (Optional) Use Docker:

   docker build -t ai-health-backend .
   docker run -p 5000:5000 ai-health-backend

Frontend

• Location: frontend/
• Tech: React, Vite, Sass, Framer-motion
• Features:
  • User dashboard, login, health assessment
  • Responsive UI
• Key files:
  • src/: Main source code
  • App.jsx, components/, scss/
  • package.json: JS dependencies

Frontend Setup

1. Navigate to frontend/
2. Install dependencies:

   npm install

3. Run development server:

   npm run dev

Mobile App (a mere UI not a working Application)

• Location: mobile/my-app/
• Tech: React Native, Expo
• Features:
  • Mobile health assistant
  • Native UI components
• Key files:
  • app/, components/, hooks/, assets/
  • package.json: JS dependencies

Mobile Setup

1. Navigate to mobile/my-app/
2. Install dependencies:

   npm install

3. Start Expo server:

   npm start

4. Run on device/emulator:

   npm run android   # For Android(UI not a working Application)
   npm run ios       # For iOS(UI not a working Application)
   npm run web       # For web

Features

• AI-powered health chat and assessment
• Symptom prediction and analysis
• User authentication
• Voice transcription
• Dashboard and reports

Requirements

• Python 3.12+
• Node.js 18+
• npm 9+
• Docker (optional)

#CAVISTA 2026 HACKATHON DASHBOARD

HEALTH SHIELD DASHBOARD REPORT

Empowering Early Detection Through Predictive Analytics

Introduction

 The Health Shield Dashboard is an automated disease-symptom intelligence system designed to support early detection and awareness. This solution 

focuses on individuals who are already familiar with a disease they frequently experience but may not clearly remember all associated symptoms. The dashboard provides an interactive, user-friendly environment where selecting any disease dynamically updates visual elements to assist in symptom verification and confirmation.

Dashboard Objective

The primary goal of this automation system is to:

-Improve early disease awareness
-Help users confirm symptoms quickly
-Provide visual identification support
-Enable gender-based disease pattern analysis
-Present structured symptom information for deeper verification

This system currently supports 20 distinct diseases, all integrated into an interactive analytics framework.

Automation Functionality

The dashboard is fully dynamic and automated:

Disease Selection Automation

When a user selects a disease (from the bar chart, slicer, or table): The corresponding disease image automatically appears. The Top 3 most common symptoms are instantly displayed. The detailed symptom table updates to show additional symptoms for deeper confirmation. This allows users to: Quickly compare their physical condition with the disease image. Verify the most critical symptoms immediately. Review additional symptoms in the table for further validation.

Key Insights

-The system currently automates detection across 20 distinct diseases. -Male individuals appear to have a higher prevalence rate across the dataset. -The dynamic filtering allows real-time disease intelligence. -The integration of image recognition with symptom ranking improves user confidence in early detection.

Predictive Analytics Value

This dashboard goes beyond static reporting by:

-Automating disease-symptom mapping
-Enabling interactive filtering
-Supporting drill-down analysis
-Structuring medical information into actionable insights

It acts as a lightweight decision-support system that enhances personal health awareness.