Could you share the model trained for subtask2?

[huangmozhilv]

Great job and thanks for the kindness to share the code and baseline models! We have a dataset consisting of multi-phase contrast enhanced abdominal CT scans, and would like to segment the livers and spleens. The model trained for subtask2, which was trained with data from multi-phases, multi-vendor and multi-centers, should perform well for our dataset.

The models in BaiduNetDisk were named 'Task311_OrganSparse5', 'Task312_OrganSparse30', 'Task313_OrganSparse15', what are their differences? According to their pkl files, the training files seem to be the 41 MSD spleen cases ('Task314_SpleenSparse15' ). Thus they don't seem to be the mdoels trained for subtask2.

Could you make the model trained subtask2 public? Thank you very much.

[JunMa11]

the models you mentioned were trained with different label ratios. Please look at the paper for more details.

Which subtask2 are you looking for? e.g., which benchmark?

[huangmozhilv]

I am looking for the model of subtask 2 for the Fully supervised abdominal organ segmentation benchmark. The training set consist of 3 datasets: MSD Pan. Plus (281), LiTS Plus (40), KiTS Plus (40).

[JunMa11]

Here you go (T122) 链接：https://pan.baidu.com/s/1BNZn4jkgbRWzIP7XXL6Tcg?pwd=2022 提取码：2022

BTW, docker would be a more convenient way to use the trained model. You do not need to set up the environment again.

[huangmozhilv]

Thank you very much. I'm not familiar with docker. Will try it later.

[JunMa11]

It will save your life (time) to use the dockerized models without configuring the environment.

You only need to install docker and nvidia-docker. Then, you can run the inference with one-line command.

Here is a good tutorial on docker https://nbviewer.org/github/ericspod/ContainersForCollaboration/blob/master/ContainersForCollaboration.ipynb

[huangmozhilv]

Thanks so much. You are so nice to share the tutorial.