https://github.com/rinat-akhmetov/ultrasound-lens

Converts ultrasound DICOM images to JPG and runs AI (LLM) analysis of the resulting images, summarizing findings in a streamlined workflow.
https://github.com/rinat-akhmetov/ultrasound-lens

anthropic dicom image-processing llm medical-imaging streamlit ultrasound

Last synced: 23 days ago
JSON representation

Converts ultrasound DICOM images to JPG and runs AI (LLM) analysis of the resulting images, summarizing findings in a streamlined workflow.

• Host: GitHub
• URL: https://github.com/rinat-akhmetov/ultrasound-lens
• Owner: rinat-akhmetov
• License: mit
• Created: 2025-01-06T17:32:58.000Z (28 days ago)
• Default Branch: main
• Last Pushed: 2025-01-06T17:40:16.000Z (28 days ago)
• Last Synced: 2025-01-06T18:45:27.142Z (28 days ago)
• Topics: anthropic, dicom, image-processing, llm, medical-imaging, streamlit, ultrasound
• Language: Python
• Homepage:
• Size: 43.9 KB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Ultrasound-Lens

## Overview

Converts ultrasound DICOM images to JPG and runs AI (LLM) analysis of the resulting images, summarizing findings in a streamlined workflow

## Installation

1. Install Python dependencies:

   - pydicom, numpy, Pillow, anthropic, streamlit, tqdm

2. Make sure ANTHROPIC_API_KEY is set in your environment.

## Using uv

You can also use uv to manage or run this project. For instance:

1. Install uv.

2. Run uv with the main.py in this directory.

## Usage

1. Run:

   - python main.py

2. Enter the DICOM folder path in the UI.

3. Click "Process" to convert and analyze images.

## Additional Information

- Images are converted and saved in the "ultrasound_images" folder by default.

- You can also run Streamlit directly with:  

  streamlit run main.py