Skip to content

Medical SAM 2: Segment Medical Images As Video Via Segment Anything Model 2

License

Notifications You must be signed in to change notification settings

rabiaedayilmaz/Medical-SAM2

 
 

Repository files navigation

● Medical SAM 2: Segment Medical Images As Video Via Segment Anything Model 2

Discord License

Medical SAM 2, or say MedSAM-2, is an advanced segmentation model that utilizes the SAM 2 framework to address both 2D and 3D medical image segmentation tasks. This method is elaborated on the paper Medical SAM 2: Segment Medical Images As Video Via Segment Anything Model 2.

🔥 A Quick Overview

🩻 3D Abdomen Segmentation Visualisation

🧐 Requirement

Install the environment:

conda env create -f environment.yml

conda activate medsam2

You can download SAM2 checkpoint from checkpoints folder:

bash download_ckpts.sh

Further Note: We tested on the following system environment and you may have to handle some issue due to system difference.

Operating System: Ubuntu 22.04
Conda Version: 23.7.4
Python Version: 3.12.4

We released our pretrain weight here

🎯 Example Cases

Download REFUGE or BCTV or your own dataset and put in the data folder, create the folder if it does not exist ⚒️

2D case - REFUGE Optic-cup Segmentation from Fundus Images

Step1: Download pre-processed REFUGE dataset manually from here, or using command lines:

wget https://huggingface.co/datasets/jiayuanz3/REFUGE/resolve/main/REFUGE.zip

unzip REFUGE.zip

Step2: Run the training and validation by:

python train_2d.py -net sam2 -exp_name REFUGE_MedSAM2 -vis 1 -sam_ckpt ./checkpoints/sam2_hiera_small.pt -sam_config sam2_hiera_s -image_size 1024 -out_size 1024 -b 4 -val_freq 1 -dataset REFUGE -data_path ./data/REFUGE

3D case - Abdominal Multiple Organs Segmentation

Step1: Download pre-processed BTCV dataset manually from here, or using command lines:

wget https://huggingface.co/datasets/jiayuanz3/btcv/resolve/main/btcv.zip

unzip btcv.zip

Step2: Run the training and validation by:

python train_3d.py -net sam2 -exp_name BTCV_MedSAM2 -sam_ckpt ./checkpoints/sam2_hiera_small.pt -sam_config sam2_hiera_s -image_size 1024 -val_freq 1 -prompt bbox -prompt_freq 2 -dataset btcv -data_path ./data/btcv

🚨 News

  • 24-08-05. Our Medical SAM 2 paper ranked #1 Paper of the day collected by AK on Hugging Face 🤗
  • 24-08-05. Update 3D example details and pre-processed BTCV dataset download link 🔗
  • 24-08-05. Update 2D example details and pre-processed REFUGE dataset download link 🔗
  • 24-08-05. Our Medical SAM 2 paper was available online 🥳
  • 24-08-05. Our Medical SAM 2 code was available on Github 🥳
  • 24-07-30. The SAM 2 model was released 🤩

📝 Cite

@misc{zhu_medical_2024,
   title={Medical SAM 2: Segment medical images as video via Segment Anything Model 2},
   author={Jiayuan Zhu and Yunli Qi and Junde Wu},
   year = {2024},
   eprint={2408.00874},
   archivePrefix={arXiv},
   primaryClass={cs.CV}
}

About

Medical SAM 2: Segment Medical Images As Video Via Segment Anything Model 2

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 93.0%
  • Python 6.8%
  • Other 0.2%