how to convert detected nodule coordinates back to original scanning coordinates?

[galactica147]

Hi @lfz

Can you please advise how to convert the nodule detection results (coordinates in .pbb) files back to the original coordinate system in the scan? For example, all luna16 data has been pre-processed and normalized, so if we want to test on luna16, how do we get the corresponding coordinates in the original system? Any ideas? I assume this reverse transform would be scan-dependent.

Thanks!

[iridescent303]

Have you found the way to convert?

[lfz]

Sorry about the late reply, you need to store the crop origin and spacing information at the preprocessing stage, which could be found in the savenpy_luna function

Besides, please refer to data_classifier to see how to convert pbb into the bounding box you need.

BTW, I know you guys are preparing for the Tianchi competition, good luck. Do not just ask for help, you guys should share thoughts with each other more frequently to get better results together.

[iridescent303]

Thank you very much!

[teaghan]

anyone have a code edit to save this information, make the conversion for the DSB data during the Test stage, and map the nodules back to the original files? Or even just for the luna16 data to give an example? Thanks!

[zhaifly]

may this script can help you! convert.txt

to gen the 'lung_segments.json' use this script states.txt

this is the "lung_segments.json" i created luna_segments.txt

[bijoygit]

Has anybody able to find the slice number of the detected nodule. Need help

[wjx2]

Hi, In the luna16 evaluation scripts, it seems the prediction probability should be in the range of 0-1, but my prediction probability is not in the range of 0-1. How can I solve this problem?