search

zzzqzhou / RAM-DSIR

[ECCV'22] Generalizable Medical Image Segmentation via Random Amplitude Mixup and Domain-Specific Image Restoration
37 stars 3 forks source link

prostate dataset #2

Closed H-CODE6 closed 1 year ago

H-CODE6 commented 2 years ago

Hello, the author. After downloading your public prostate dataset, it is found that it includes Domain1-Domain6, 'ISBI' and 'ISBI 1.5 ',' I2CVB ',' UCL ',' BIDMC ',' HK 'folders. After testing, I found that Domain1-Domain6 folders were used in training, but only' ISBI 'and' ISBI 'could be used in testing 1.5 ',' I2CVB ',' UCL ',' BIDMC ',' HK 'folders, but I am not clear about the Domain1-Domain6 folder and the' ISBI ',' ISBI_ Please inform us of the correspondence between the folders 1.5 ',' I2CVB ',' UCL ',' BIDMC 'and' HK 'and look forward to your reply.

zzzqzhou commented 2 years ago

Hi. Thanks for your interest in our work. We have a test domain list in 'code/test_prostate_volume.py'. The list index is also the domain id. Which means ISBI is Domain 1, ISBI_1.5 is Domain 2, I2CVB is Domain 3, UCL is Domain 4, BIDMC is domain 5 and HK is Domain 6. https://github.com/zzzqzhou/RAM-DSIR/blob/cb4aa2b8a52ebf8b50b6afbd9a122554602161a3/code/test_prostate_volume.py#L167

Images in ISBI, ISBI_1.5, etc. are all 3D volumes. We preprocessed these images into 2D slices (Domain 1 - 6) for training. When testing, we use 3D volumes.

H-CODE6 commented 2 years ago

Thank you very much for your reply. Your answer has solved my confusion. If it is convenient, can I open the code for converting 3D to 2D? This code is relatively complete. I look forward to your reply.

H-CODE6 commented 2 years ago

Opening the prostate data set preprocessing code can deepen our understanding of the overall code, and we look forward to your reply.