Is there a way to interact programmatically with Medical Image Labeler?

Question

Nicola Dall'Osto 2024-4-12

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2106371-is-there-a-way-to-interact-programmatically-with-medical-image-labeler

回答： Cris LaPierre 2024-4-14

I know that there are some functions that consent to interact with Medical Image Labeler app of Medical Imaging Toolbox, but the operations allowed are very few. In particular I'm looking for a method to open nifti images, segment and save the labels by setting up a script or something else.

Is it feasible?

4 个评论
显示 2更早的评论隐藏 2更早的评论

Cris LaPierre 2024-4-13

What do you need the script to do? Load the images into the app?

Your session in the medical image labeler app will automatically save your labels to a Nifti file in the session folder.

Nicola Dall'Osto 2024-4-14

编辑：Nicola Dall'Osto 2024-4-14

Yes, I need to automatically upload images into the app. In particular, I am facing a problem in this step. I've described the issue in detail here.

I know that the labels are saved in a .mat file, not nifti. Anyway, it is not a problem as long as the labels are automatically saved in the session folder.

To upload the unlabeled images I use groundTruthMedical(dataSource,labelDefs,labelData), but the user has to set a new session. Ideally, I want that running that command, a new session starts without asking the users to choose the path of the new session folder. Maybe I am asking too much.

Finally, I really need that the labels are saved with the user's ID or the user's name in order to be able to know who has segmented each image, since more people will segment different images.

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Cris LaPierre 2024-4-14

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2106371-is-there-a-way-to-interact-programmatically-with-medical-image-labeler#answer_1441056

I have responded in your other question.

The Label Definitions are saved to a mat file, along with paths to the Data Source and Labeled Data. Your labeled volume is saved as a Nifti file in the LabelData folder in your session folder.