This project provides a pure frontend simulation of multi-doctor AI medical consultations. It addresses the need for collaborative AI diagnosis by allowing multiple Large Language Models (LLMs) to act as different medical specialists, discussing and evaluating a patient's case. The system is beneficial for researchers, developers, and power users exploring AI-driven medical diagnostics without requiring backend infrastructure.

How It Works

The system is built entirely on the frontend using Vue 3, Pinia for state management, and Ant Design Vue for the UI. It integrates with various LLM providers (OpenAI, Anthropic, Gemini, SiliconFlow, Modaa) by calling their APIs directly from the browser. Multiple LLM instances are configured as distinct "doctors" who collaboratively discuss a medical case, taking turns to present their findings. After each round, doctors automatically evaluate each other's diagnoses, and those receiving the most negative feedback are eliminated, converging towards a consensus diagnosis.

Quick Start & Requirements

• Primary install/run: Clone the repository, cd ai-medical-consultation-panel, then run pnpm install (or npm install) followed by pnpm dev (or npm run dev) to start the development server.
• Prerequisites: Node.js >= 16.0.0, pnpm >= 9.0.0 (recommended) or npm. Crucially, users must obtain and configure API keys for the desired LLM providers.
• Links: Project preview available at https://dragonchencl.github.io/ai-doctor/.

Highlighted Details

• Supports multiple LLM providers and models, including OpenAI, Anthropic, Gemini, SiliconFlow, and Modaa.
• Features an automated peer-review and elimination mechanism for simulated doctors to refine diagnoses.
• Pure frontend architecture eliminates the need for a backend server, with all data stored locally in the browser's localStorage.
• Includes real-time discussion display with a typewriter effect and robust session management.

Maintenance & Community

Community links such as Discord or Slack channels are not explicitly detailed in the provided README. Contributions are welcomed via GitHub Issues and Pull Requests.

Licensing & Compatibility

The project is licensed under the MIT License, which permits commercial use and integration into closed-source projects.

Limitations & Caveats

As a pure frontend application, all configurations, including sensitive API keys, are stored in the browser's localStorage. Clearing browser data will result in the loss of all saved configurations and session history. Users are responsible for managing their API keys securely and ensuring they have access to the necessary third-party AI services.