AGNOS AI: Advanced 3D Anatomical Symptom Mapping

AGNOS AI is an interactive diagnostic platform designed to provide a spatial and technical bridge between patient symptoms and anatomical understanding. By utilizing a high-fidelity 3D interface, users can map their physical discomfort directly onto a digital human model, enabling more precise communication of symptoms than traditional text-based interfaces.

Overview & Background

The core challenge in remote healthcare or digital symptom assessment is the "descriptive gap." Patients often struggle to articulate the exact location, depth, and nature of their pain to a standard chatbot. AGNOS AI solves this by making 3D mapping the primary mode of interaction. Instead of describing a "headache," a user can precisely mark the supraorbital region or the temporal fascia, allowing for a more granular starting point for clinical analysis.

Key Features

3D Symptom Mapping

The central interface allows users to rotate, inspect, and "paint" symptomatic regions directly onto the model. This spatial data is then processed to identify specific anatomical zones, providing a technical baseline for the diagnostic assessment.

Intelligent AI Diagnostics

Once a region is marked, the integrated Agnos AI—powered by the Claude LPU™ and specialized medical prompt engineering—analyzes the mapped metadata. It provides concise, professional, and warm feedback via synchronized text and speech.

Anatomical Layering & Subtitles

To facilitate a deeper understanding, the system offers visualization toggles for muscles, internal organs, and specialized 3D subsystems like the ocular and laryngeal structures.

Ethical Considerations

AGNOS AI is built with a commitment to responsible AI deployment and patient safety:

• Clinical Guardrails: If the system detects symptoms synonymous with life-threatening conditions (e.g., severe chest pain or neurological indicators), it is programmed to prioritize an immediate recommendation for professional emergency care.
• Medical Disclaimer: The platform explicitly maintains its status as an educational and visualization tool, ensuring users understand it is not a substitute for professional medical advice.
• Bias Mitigation: The AI is prompted to consider physiological diversity—including age, gender identity, and skin tone—to ensure that feedback remains objective and unbiased across different demographics.
• Privacy First: The current architecture is designed for local processing of coordinates, minimizing the storage of sensitive personal health information (PHI) within the primary application layer.

Technical Architecture

• Logic & Rendering: React with Three.js (via React Three Fiber and Drei) for high-performance 3D visualization and decal mapping.
• Natural Language Processing: Anthropic's Claude 3.5 Sonnet API for high-accuracy, low-latency diagnostic reasoning.
• Human-Computer Interaction: Web Speech API for seamless browser-native text-to-speech output, ensuring high availability and zero-cost accessibility.
• Data Export: PDF generation module for session summaries, allowing users to take their mapped data to a healthcare provider.

Installation and Setup

To run AGNOS AI locally, ensure you have Node.js installed on your system.

1. Clone the repository:

   git clone https://github.com/yourusername/agnos-ai.git
   cd agnos-ai

2. Install dependencies:

   npm install

3. Environment Configuration: Create a .env file in the root directory and add your API keys:

   VITE_CLAUDE_API_KEY=your_claude_api_key
   VITE_ELEVENLABS_API_KEY=your_elevenlabs_api_key

   Note: The platform is configured to securely utilize the ElevenLabs ultra-low-latency eleven_turbo_v2_5 TTS model alongside native HTML5 blob audio synchronization.

4. Execute development server:

   npm run dev

Navigate to http://localhost:5173 to interact with the platform.