Adibvafa commited on
Commit
f0ada78
·
2 Parent(s): 0e1a2db bd8efea

Merge branch 'main' of github.com:bowang-lab/MedRAX

Browse files
Files changed (1) hide show
  1. README.md +13 -1
README.md CHANGED
@@ -1,8 +1,14 @@
1
  <h1 align="center">
2
  🤖 MedRAX: Medical Reasoning Agent for Chest X-ray
3
  </h1>
 
 
4
  <br>
5
 
 
 
 
 
6
  ## Abstract
7
  Chest X-rays (CXRs) play an integral role in driving critical decisions in disease management and patient care. While recent innovations have led to specialized models for various CXR interpretation tasks, these solutions often operate in isolation, limiting their practical utility in clinical practice. We present MedRAX, the first versatile AI agent that seamlessly integrates state-of-the-art CXR analysis tools and multimodal large language models into a unified framework. MedRAX dynamically leverages these models to address complex medical queries without requiring additional training. To rigorously evaluate its capabilities, we introduce ChestAgentBench, a comprehensive benchmark containing 2,500 complex medical queries across 7 diverse categories. Our experiments demonstrate that MedRAX achieves state-of-the-art performance compared to both open-source and proprietary models, representing a significant step toward the practical deployment of automated CXR interpretation systems.
8
  <br><br>
@@ -25,6 +31,7 @@ MedRAX is built on a robust technical foundation:
25
  - **X-ray Generation**: Utilizes RoentGen for synthetic CXR generation
26
  - **Utilities**: Includes DICOM processing, visualization tools, and custom plotting capabilities
27
 
 
28
  <br>
29
 
30
  ## ChestAgentBench
@@ -62,7 +69,12 @@ pip install -e .
62
  # Start the Gradio interface
63
  python main.py
64
  ```
65
- You need to setup the `model_dir` to directory where you want to download or already have the weights of above tools from Hugging Face. Comment out the tools that you do not have access to.
 
 
 
 
 
66
  Make sure to setup your OpenAI API key in `.env` file!
67
  <br><br>
68
 
 
1
  <h1 align="center">
2
  🤖 MedRAX: Medical Reasoning Agent for Chest X-ray
3
  </h1>
4
+ <p align="center"> <a href="https://arxiv.org/abs/2502.02673" target="_blank"><img src="https://img.shields.io/badge/arXiv-Paper-FF6B6B?style=for-the-badge&logo=arxiv&logoColor=white" alt="arXiv"></a> <a href="https://github.com/bowang-lab/MedRAX"><img src="https://img.shields.io/badge/GitHub-Code-4A90E2?style=for-the-badge&logo=github&logoColor=white" alt="GitHub"></a> <a href="https://huggingface.co/datasets/wanglab/chest-agent-bench"><img src="https://img.shields.io/badge/HuggingFace-Dataset-FFBF00?style=for-the-badge&logo=huggingface&logoColor=white" alt="HuggingFace Dataset"></a> </p>
5
+
6
  <br>
7
 
8
+ [![Watch the video](https://github.com/bowang-lab/MedRAX/blob/main/assets/medrax_logo.png)](https://github.com/bowang-lab/MedRAX/raw/main/assets/demo_fast.mp4)
9
+
10
+
11
+
12
  ## Abstract
13
  Chest X-rays (CXRs) play an integral role in driving critical decisions in disease management and patient care. While recent innovations have led to specialized models for various CXR interpretation tasks, these solutions often operate in isolation, limiting their practical utility in clinical practice. We present MedRAX, the first versatile AI agent that seamlessly integrates state-of-the-art CXR analysis tools and multimodal large language models into a unified framework. MedRAX dynamically leverages these models to address complex medical queries without requiring additional training. To rigorously evaluate its capabilities, we introduce ChestAgentBench, a comprehensive benchmark containing 2,500 complex medical queries across 7 diverse categories. Our experiments demonstrate that MedRAX achieves state-of-the-art performance compared to both open-source and proprietary models, representing a significant step toward the practical deployment of automated CXR interpretation systems.
14
  <br><br>
 
31
  - **X-ray Generation**: Utilizes RoentGen for synthetic CXR generation
32
  - **Utilities**: Includes DICOM processing, visualization tools, and custom plotting capabilities
33
 
34
+ Note the current version of MedRAX is experimentally released and does not support vision for GPT-4o and MedSAM. We will be integrating these shortly.
35
  <br>
36
 
37
  ## ChestAgentBench
 
69
  # Start the Gradio interface
70
  python main.py
71
  ```
72
+ or if you run into permission issues
73
+ ```bash
74
+ sudo -E env "PATH=$PATH" python main.py
75
+ ```
76
+ You need to setup the `model_dir` to directory where you want to download or already have the weights of above tools from Hugging Face.
77
+ Comment out the tools that you do not have access to.
78
  Make sure to setup your OpenAI API key in `.env` file!
79
  <br><br>
80