A Full Remote Autism Care Ecosystem
AI-powered neurodevelopmental screening, interactive therapy, and clinical intelligence.
NeuroBridge AI has evolved from a screening tool into a comprehensive Continuous Remote Care Ecosystem. Designed for caregivers and clinicians, it combines real-time computer vision, interactive therapy games, and clinical intelligence tools into a cohesive web platform to assist in early ASD screening and therapy planning.
- Frontend Application: https://neurobridge-app.onrender.com
- Backend API: https://neurobridge-api.onrender.com
(Note: Initial load may take up to 50 seconds on the free tier of Render).
A robust initial screening tool that uses the device camera to analyze facial and behavioral features.
- Real-time Face Analysis: Uses MediaPipe Face Mesh (468 landmarks) to track expressions.
- Hybrid Gaze Tracking: Iris and nose-based attention scoring to assess focus and eye contact.
- Behavioral Questionnaire Fusion: Multi-modal risk assessment combining AI metrics with standardized questions.
- Triage Classification: Provides severity classification with clinical explanations.
An interactive hub with 4 custom-built Phaser games designed specifically for neurodevelopmental therapy.
- Memory Match: Visual memory training with OpenMoji emoji cards, animations, and tiered difficulty.
- Day Builder: ADL (Activities of Daily Living) sequencing using drag-and-drop mechanics to build daily routines.
- Emotion Mirror: Webcam-based emotion matching game that prompts children to mimic expressions.
- Gaze Garden: Sustained attention training using a click-and-hold focus mechanic.
- Caregiver Co-Play: Includes note-taking functionality during games and game session replays.
A powerful dashboard designed for clinical professionals to monitor patient progress over time.
- Patient Trends & Cohort Analytics: Track engagement, therapy effectiveness, and overall progress.
- Regression Alerts: Automatically detects regressions in skills and alerts the clinician.
- Therapy Goal Management: Set, track, and update domain-based therapy goals.
- Recharts-powered Dashboards: Comprehensive line, area, bar, and radar charts.
A safe, supportive forum for caregivers to share experiences, ask questions, and build a network.
- ML Content Moderation: Integration with TensorFlow.js Toxicity model for automated moderation and profanity filtering.
- "Ask a Clinician" Section: Dedicated category with clinician verification badges.
- Community Features: Pinned posts, bookmarks, likes, nested comments, and post reporting.
Integrated video calls for remote consultations and therapy sessions.
- WebRTC Video Calls: Uses Socket.IO for signaling and Metered TURN servers for reliable real-time video.
- Appointment Management: Weekly calendar grid views for easy scheduling.
- Session Notes & Summaries: Pre-visit clinical snapshots and post-session note keeping.
Proactive tools to keep caregivers and patients engaged.
- Visual Schedule with Voice Readback: Daily routines using the Web Speech API.
- Automated Push Notifications: Web Push and VAPID powered reminders for therapy and appointments.
- Email Reports: Weekly progress email digests using Nodemailer.
- Automated Scheduler: Cron jobs for daily reminders, notification cleanup, and streak tracking.
This project is built with a modern, decoupled architecture capable of seamless scaling:
| Layer | Technologies |
|---|---|
| Frontend | React 19, Vite 7, Tailwind CSS 4, Recharts, Phaser 3, Framer Motion |
| Backend | Node.js 22, Express 5, Socket.IO, Nodemailer, web-push |
| Database | SQLite (via better-sqlite3) — Designed to be PostgreSQL migration-ready |
| AI / ML | MediaPipe Face Mesh, TensorFlow.js Toxicity Model, Web Speech API |
| Testing | Playwright (E2E Suites) |
- Sensory-Safe / Calm Mode: Disables harsh animations, mutes gradients, and reduces cognitive load at the flip of a switch.
- Age-Adaptive UI: Automatically scales typography and spacing based on user age profiles.
- API Versioning Prefix: Future-proofed under
/api/v1with a centralized API configuration utility.
- Node.js 22.0.0 or higher
- npm 10.0.0 or higher
-
Clone the Repository
git clone https://github.com/harshitworkmain/Neurobridge.git cd Neurobridge git checkout neurobridge-v2 -
Install Dependencies
npm install
-
Environment Configuration
cp .env.example .env
(The default variables are sufficient for local development. If you wish to test Email or Push Notifications, you will need to add SMTP credentials and VAPID keys to the
.envfile). -
Start the Backend Server (Port 3001)
node server/index.js
-
Start the Frontend Application (Port 5173)
# Open a new terminal instance npm run dev -
View the Application Open your browser and navigate to http://localhost:5173.
This project is part of academic research. All rights reserved.
⚠️ Disclaimer: NeuroBridge is a continuous remote care and screening support tool. It is not a diagnostic instrument. All screening and assessment results are intended to assist caregivers and should be reviewed by a qualified healthcare professional.