junma11 / sota-medseg Goto Github PK

Would you please consider adding TorchIO to the list of open-source tools?

Hi Jun,

Thank you for the great work. I trust this repository is indeed valuable for the community!

Upon viewing the teaser image, I noticed that there is limited work focusing on endoscopy images, particularly on Early Esophageal Cancer (EEC). Therefore, I would like to introduce our recent work published in IEEE Transactions on Image Processing on this topic:

arxiv version: Single-Image-Based Deep Learning for Segmentation of Early Esophageal Cancer Lesions
IEEE version: coming soon

In this work, we have collected EEC datasets with the assistance of doctors from West China Hospital (Huaxi Hospital). Additionally, we propose a zero-shot strategy to achieve lesion segmentation.

I would greatly appreciate it if you could review our paper and consider including it in this repository.
Thank you very much!

By the way, I like your MedSAM and your comment on the DDNM repository issue was really helpful to me, as I am also investigating the application of generative super-resolution on medical images.

Best regards,

Haipeng Li
PhD Student at the University of Electronic Science and Technology of China

Hello, would you please consider adding MedicalZooPytorch on the open source tools list?
Link: https://github.com/black0017/MedicalZooPytorch

Hi, @JunMa11

I am very interested in this job，will the excellent methods be open for us to learn? And I have another question to ask, I have emailed the signed data confidentiality form for three weeks, but I still do not have permission to download the data, so I want to ask how long did you get the data.

Thank you very much!

Hiiii Jun, thanks for your nice leaderboard collecting repo!
Here I have a question for LiTS 2017 dataset performance: https://competitions.codalab.org/competitions/17094#results
I noticed in your paper "How Distance Transform Maps Boost Segmentation CNNs: An Empirical Study" you used this dataset for loss comparison. I'm so confused why the performance could vary so far, just like on leaderboard, the liver tumour segmentation result could be up to 0.7x and the results mentioned in your paper is 0.5x. Thougn there may exist some difference on dataset split and processing.

I really want to use some template dataset processing codes for this dataset to get 0.7x on baselines like V-Net. Could your give some more guidance? Thanks a lot in advance!

Hi JunMa11,
Do you know if can download asoca data at this time，I‘ve sent email and didn't get reply. Could you share your asoca data.Thanks really a lot

您好，对于2018 MICCAI Medical Segmentation Decathlon的Pancreas & Tumor结果中，Qihang Yu et al. (paper)一文我看到了他们在table3中汇报了80.37/56.36的结果，说否说胰腺癌目前的SOTA分割结果为56.36呢？还是说以table1中汇报的54.41为准？