Skip to content

Commit 67f24db

Browse files
authored
Update README.md
1 parent 0dcffb3 commit 67f24db

File tree

1 file changed

+41
-8
lines changed

1 file changed

+41
-8
lines changed

README.md

+41-8
Original file line numberDiff line numberDiff line change
@@ -1,9 +1,20 @@
1-
# MediNotes
2-
The Capstone Research project is still in progress. We will release the results gradually, part by part. </br>
3-
You can try the fine-tuned Llama3 that can convert medical dialogues to SOAP notes now.
1+
# Research Capstone Project: MediNotes
2+
MediNotes: SOAP Note Generation through Ambient Listening, Large Language Model Fine-Tuning, and RAG
3+
- MediNotes was awarded “Best in Show” as one of the Top Capstone Projects at the University of Chicago showcase.
4+
- This project was a collaboration with UChicago Medicine to advance healthcare AI.
5+
- Building on groundbreaking research from the Microsoft AI team published in Nature, we developed an innovative framework designed to streamline medical documentation and the consultation process, with the goal of alleviating physician burnout.
6+
- By combining cutting-edge technologies like ambient listening, large language model fine-tuning, and retrieval-augmented generation (RAG), MediNotes represents a significant step forward in optimizing healthcare workflows and improving physician efficiency.
47

8+
## Demo Preview:
9+
<a href="http://www.youtube.com/watch?feature=player_embedded&v=QmWfvFdQc08
10+
" target="_blank"><img src="http://img.youtube.com/vi/QmWfvFdQc08/0.jpg"
11+
alt="IMAGE ALT TEXT HERE" width="640" height="360" border="10" /></a>
512

6-
## Setup
13+
## Live Demo
14+
This may contain error for preview for raw demo because you may need GPU more to 4gb to run the model: <br>
15+
https://medinotes-llm.streamlit.app/search
16+
17+
## Code Setup
718
1. Download the requirements.txt and app.py file.
819
2. Install the necessary libraries.
920

@@ -16,9 +27,31 @@ You can try the fine-tuned Llama3 that can convert medical dialogues to SOAP not
1627
```python
1728
python -m streamlit run app.py
1829
```
30+
31+
## Model:
32+
You can try the fine-tuned model yourself, which converts medical dialogues into SOAP notes:
33+
34+
Full 4-bit quantized model:
35+
https://huggingface.co/Yuki20/llama3_8b_aci_3e_full4b
36+
37+
Full model:
38+
https://huggingface.co/Yuki20/llama3_8b_aci_3e_full
39+
40+
Adapter only with Unsloth: <br>
41+
https://huggingface.co/Yuki20/mistral_7b_aci_3e <br>
42+
https://huggingface.co/Yuki20/llama3_8b_aci_5e <br>
43+
https://huggingface.co/Yuki20/llama3_8b_aci <br>
1944

20-
4. Click the generated URL to launch the ChatBot.
21-
<img src="https://github.com/yuki-2025/MediNotes/assets/159591455/d2f64ee2-8bce-45f6-9066-03b7921d8e82" alt="Untitled design" width="300" height="150">
45+
## Citation
46+
If you find Medinotes useful in your research or applications, please kindly cite:
47+
```
48+
49+
@inproceedings{leong2024medinotes,
50+
title={{MediNotes}: A Generative AI Framework for Medical Note Generation},
51+
author={Leong, HY and Gao, YF and Ji, S and Kalaycioglu, Bora and Pamuksuz, Uktu},
52+
journal={arXiv preprint arXiv:2410.01841},
53+
year={2024} }
54+
```
2255

23-
## Preview
24-
<img src="https://github.com/yuki-2025/MediNotes/assets/159591455/30eebc8a-1675-4bfd-bef9-a04f646b6b64" alt="Untitled design" width="500" height="500">
56+
## Acknowledgements
57+
You may refer to related work that serves as foundations for our framework and code repository, Aci-bench. Thanks for their wonderful works.

0 commit comments

Comments
 (0)