Mamba-Sea: A Mamba-based Framework with Global-to-Local Sequence Augmentation for Generalizable Medical Image Segmentation
Mamba-Sea has been accepted by IEEE TMI on April 23, 2025. 【Paper】
The core part of Mamba-Sea's code is in Network1 and Network2.By using these two files as a backbone, you can easily incorporate our code into your tasks. The complete part of our code will soon be released by May 2025.
-
'./dataset/fundus/' or './dataset/ProstateSlice/' or './dataset/skin/'
-Domain1
- train
- images
- .png
- masks
- .png
- images
- test
- images
- .png
- masks
- .png ......
- images
- train
The weights of the pre-trained VMamba could be downloaded Google Drive. After that, the pre-trained weights should be stored in './pretrained_weights/'.
Thanks a lot for the valuable work of VM-UNet and VMamba.
Download dataset Fundus (Provided by DoFE).
Download dataset Prostate (Originally Provided by SAML and RAM-DSIR).
Download dataset ISIC2018 and PH2. (Following ESP-MedSAM).
Email: chengzihan@sjtu.edu.cn or czh@smail.nju.edu.cn, each one is OK.