DoctorAI — Vital Image Analytics

DoctorAI (VitalImage Analytics) is a Streamlit-based AI medical image analysis assistant that provides structured assessments of uploaded medical images using generative AI. Its goal is to help users (clinicians, researchers, or general users) explore potential anomalies and health insights in medical images through automated interpretation powered by Google’s Gemini GenAI.
Disclaimer: This tool provides informative suggestions only and is not a substitute for professional medical diagnosis or treatment.

What This Project Does

DoctorAI lets a user:

• Upload a medical image (PNG/JPG)
  
• Automatically generate detailed medical interpretation
  
• Review findings, recommendations, and suggested next steps
  

The app leverages an AI model configured for medical image analysis context and structures responses to identify abnormalities, document observations, and offer high-level recommendations — all accompanied by a clinical disclaimer.

You can interact with the app at:
https://doctoraiassist.streamlit.app/

How It Works

The application is built with:

• Streamlit — for UI and uploads
  
• Google Gemini AI — for generative image interpretation
  
• Python scripting — to process uploaded files and organize AI prompts
  

### The core pipeline: - User uploads a medical image via the Streamlit UI
- The system constructs a structured AI prompt geared toward image analysis
- The configured Gemini model evaluates the image and returns findings
- The app displays AI analysis in a clear, structured text output

### The prompt instructs the model to:
- Analyze the image for anomalies or conditions
- Summarize findings in organized text
- Provide recommendations when possible
- Include a medical disclaimer to emphasize clinical consultation

This makes the app suitable for informational analysis and education, not for actual diagnosis.

Repository Structure

saamm/DoctorAI/
├── .devcontainer/           # Development environment configurations<br>
├── .idea/                   # IDE settings<br>
├── README.md                # This documentation<br>
├── requirements.txt         # Python dependencies<br>
├── streamlit_app.py         # Main Streamlit application logic<br>
├── Untitled.ipynb           # Example or exploratory notebook<br>

Getting Started

1 Clone the repo:

git clone https://github.com/saamm/DoctorAI.git
cd DoctorAI

2. Install dependencies:

pip install -r requirements.txt

3. Set your Google Generative AI API key:

export API_KEY="YOUR_GOOGLE_AI_API_KEY"

4. Run the app locally:

streamlit run streamlit_app.py

Usage Guide

1. Open the Streamlit app in your browser
2. Click “Choose file” to upload a medical image
3. Press Generate the Analysis
4. Review the structured AI output
5. Use results as supplementary insights, not medical advice

Example Output Flow

The app typically generates:
Observation Summary: What the model sees
Key Findings: Potential anomalies or notable patterns
Recommendations: Follow-up actions to consider
Mandatory Disclaimer: Advising professional consultation
This structured feedback helps users interpret complex scans even if they don’t have a medical background.

Limitations & Ethics

This model’s output is not clinical diagnosis
Always consult a qualified health professional before acting
The app may misinterpret poor quality images or atypical cases

Contributions

Want to improve DoctorAI?
Add support for more medical file formats (DICOM, TIFF)
Integrate visual overlays highlighting detected features
Build a session history for longitudinal analyses
Feel free to fork, update, and submit a pull request!