AI-Powered Mental Health Screening Platform
A comprehensive, professional-grade mental health screening platform that combines clinically validated assessment tools with AI-powered analysis using GPT-4o and DeepSeek-V3 via GitHub Models.
🚀 Visit MindScreen AI - Experience the live platform
Deploy your own instance with one click:
Note: For AI features, add your GitHub Models API key as
VITE_GITHUB_TOKENin environment variables.
- 🔬 PHQ-9 Depression Screening: Fully implemented, clinically validated assessment
- 🤖 AI-Powered Analysis: Advanced interpretation using GPT-4o and DeepSeek-V3 models
- 📊 Instant Results: Real-time scoring with severity classification
- 📄 PDF Reports: Downloadable comprehensive assessment reports
- 🔒 Privacy-First: No data stored on servers, HIPAA-conscious design
- 📱 Responsive Design: Modern, accessible UI that works on all devices
- 🌙 Dark Mode: Beautiful light and dark themes
- 🌍 Location Services: Smart location detection for personalized resources
- Frontend: React 19 + TypeScript + Vite
- Styling: Tailwind CSS 4 + ShadCN UI components
- State Management: Zustand
- Forms: React Hook Form + Zod validation
- Routing: React Router DOM
- Icons: Lucide React
- AI: GPT-4o (Primary) + DeepSeek-V3 (Backup) via GitHub Models API
- PDF Generation: React PDF Renderer
- Deployment: Vercel with automatic CI/CD
| Condition | Tool | Questions | Duration | Status |
|---|---|---|---|---|
| Depression | PHQ-9 | 9 | 3-5 min | ✅ Live |
| Anxiety | GAD-7 | 7 | 2-4 min | 🚧 In Development |
| ADHD | ASRS | 18 | 5-8 min | 📋 Planned |
| Autism | AQ-10 | 10 | 3-5 min | 📋 Planned |
| PTSD | PCL-5 | 20 | 5-10 min | 📋 Planned |
| OCD | OCI-R | 18 | 4-6 min | 📋 Planned |
| Bipolar | MDQ | 15 | 4-6 min | 📋 Planned |
| General Distress | K10 | 10 | 3-5 min | 📋 Planned |
- Node.js 18+
- npm or yarn
- GitHub Models API key (for AI features)
-
Clone the repository
git clone https://github.com/mishwani7/MindScreen-AI.git cd MindScreen-AI -
Install dependencies
npm install
-
Configure AI Integration (Optional but Recommended)
cp .env.example .env # Add your GitHub Models API key to .env -
Start development server
npm run dev
-
Open in browser
http://localhost:5173
MindScreen AI uses GPT-4o (Primary) and DeepSeek-V3 (Backup) via GitHub Models to provide:
- Personalized Analysis: Deep interpretation of assessment results
- Custom Recommendations: Tailored suggestions based on response patterns
- Risk Assessment: AI-identified risk and protective factors
- Professional Referral Guidance: Smart recommendations for seeking help
- Support Resources: Curated resources based on individual needs
- Multi-Model Reliability: Automatic fallback between AI models for consistent service
Note: AI-powered features require a GitHub Models API key. Without it, the platform still functions using a demo mode that provides basic, static analysis and insights.
- Step-by-step question navigation with progress tracking
- Real-time scoring with clinical severity classification
- Comprehensive results page with detailed interpretation
- AI-powered personalized insights (when configured)
- Risk assessment and safety recommendations
- PDF report generation for healthcare providers
- Location-based resource recommendations
- Based on the validated PHQ-9 assessment tool
- Implements standard clinical scoring algorithms
- Provides severity classifications: Minimal, Mild, Moderate, Moderately Severe, Severe
- Includes appropriate disclaimers and safety information
src/
├── components/ # Reusable UI components
│ ├── ui/ # ShadCN UI components
│ ├── Layout.tsx # Main layout component
│ ├── CleanResults.tsx # Results display component
│ └── PDFReport.tsx # PDF generation component
├── pages/ # Page components
│ ├── HomePage.tsx # Landing page
│ ├── ScreenersPage.tsx # Screeners overview
│ ├── PHQ9ScreenerPage.tsx # PHQ-9 implementation
│ └── AboutPage.tsx # About page
├── services/ # Business logic
│ ├── aiService.ts # AI integration
│ ├── PHQ9Processor.ts # PHQ-9 scoring logic
│ └── locationService.ts # Location services
├── store/ # Zustand state management
│ └── app-store.ts
├── data/ # Static data and configurations
│ ├── phq9.ts # PHQ-9 questions
│ └── countriesAndCities.ts # Location data
└── types/ # TypeScript type definitions
└── assessment.ts
Create a .env file:
VITE_GITHUB_TOKEN=your_github_models_api_key_here- Get a GitHub Models API key from GitHub Models
- Add the key to your
.envfile - The platform will automatically enable AI features when configured
This project is configured for automatic deployment on Vercel:
- Go to Vercel
- Connect GitHub: Import your GitHub repository
mishwani7/MindScreen-AI - Configure Environment Variables in Vercel Dashboard:
VITE_GITHUB_TOKEN= Your GitHub Models API key
- Deploy: Vercel will automatically build and deploy
AI Configuration: The platform uses GPT-4o as primary and DeepSeek-V3 as backup for reliability.
✅ Auto-Deploy: Every push to main branch triggers automatic deployment.
- No Server Storage: All data processed locally in browser
- Encrypted Transmission: Secure communication with AI services
- HIPAA-Conscious: Designed with healthcare privacy in mind
- Open Source: Full transparency in data handling
- No Tracking: No analytics or user behavior tracking
- Purpose: Depression screening and severity measurement
- Validation: Extensively validated in primary care and psychiatric settings
- Scoring: 0-27 scale with established severity thresholds
- Clinical Use: Widely used by healthcare professionals worldwide
All planned screening tools are based on established clinical assessments with published validation studies and are commonly used in healthcare settings.
This platform is for educational and informational purposes only. It is not intended to replace professional medical advice, diagnosis, or treatment. If you're experiencing mental health concerns, please consult with a qualified healthcare provider or mental health professional.
In case of emergency or thoughts of self-harm, please contact:
- Emergency Services: 911 (US) or your local emergency number
- Crisis Text Line: Text HOME to 741741
- National Suicide Prevention Lifeline: 988
We welcome contributions from developers, mental health professionals, and anyone passionate about improving mental health accessibility through technology!
- Node.js 18+ and npm
- Git
- Basic knowledge of React, TypeScript, and mental health screening principles
-
Fork and clone the repository
git clone https://github.com/mishwani7/MindScreen-AI.git cd MindScreen-AI -
Install dependencies
npm install
-
Set up environment variables
cp .env.example .env # Add your GitHub Models API key for AI features -
Start development server
npm run dev
- Implement additional validated assessment tools (GAD-7, ASRS, etc.)
- Follow existing patterns in
/src/pages/and/src/services/ - Ensure clinical accuracy and proper scoring
- Improve AI prompts and response processing
- Add new AI-powered features
- Optimize AI integration patterns
- Enhance accessibility features
- Improve responsive design
- Add new UI components following ShadCN patterns
- Use TypeScript strict mode with proper typing
- Follow existing interface patterns in
/src/types/ - Use functional components with hooks
- Follow ShadCN UI component patterns
- Maintain accessibility standards (WCAG 2.1)
Required file structure:
src/
├── pages/NewScreenerPage.tsx # Main component
├── services/NewScreenerProcessor.ts # Scoring logic
├── data/newScreener.ts # Questions data
└── types/assessment.ts # Type definitions
- Use only validated, published assessment tools
- Implement exact scoring algorithms from clinical literature
- Include appropriate disclaimers and safety information
- Test against known case examples
- Create an issue for discussion before starting work
- Create a feature branch from main
- Follow naming convention:
feature/screener-name,fix/issue-description - Submit a pull request with clear description
- Include clear description of changes
- Add screenshots for UI changes
- Confirm clinical accuracy for new screeners
- Ensure tests pass and no errors in console
- Never store sensitive user data
- Process everything client-side when possible
- Use secure transmission for AI requests
- Follow HIPAA-conscious design principles
- Never commit API keys to repository
- GitHub Issues: For bugs and feature requests
- GitHub Discussions: For questions and general discussion
- Email: abuzarmishwani742@gmail.com for sensitive matters
- PHQ-9 Depression Screening
- AI Integration with GPT-4o and DeepSeek-V3
- Modern UI with ShadCN components
- PDF Report generation
- Location-based services
- GAD-7 Anxiety Assessment
- Enhanced AI analysis
- User experience improvements
- Complete suite of 8 mental health screeners
- Advanced analytics and insights
- Multi-language support
- Healthcare provider dashboard
This project is licensed under the MIT License - see the LICENSE file for details.
If you find MindScreen AI helpful and would like to support its development, consider buying me a coffee! Your support helps:
- 🔧 Maintain and improve existing features
- 🆕 Develop new screening tools (GAD-7, ASRS, etc.)
- 🤖 Enhance AI capabilities and analysis accuracy
- 🌍 Add multi-language support for global accessibility
- 📚 Create educational resources and documentation
- 🔒 Ensure long-term sustainability of the platform
Every contribution, no matter the size, makes a meaningful difference in advancing mental health accessibility through technology.
- Lead Developer: Abu Zar Mishwani
- Clinical assessment tools developed by mental health professionals
- ShadCN for the beautiful UI component library
- GPT-4o and DeepSeek teams for the advanced AI models
- Open source community for the amazing tools and libraries
- GitHub Issues: Report bugs or request features
- Email: abuzarmishwani742@gmail.com
Built with ❤️ for mental health awareness and accessibility
Making mental health screening accessible to everyone