Jan Valosek
Jan Valosek
Related: https://github.com/neuropoly/idea-projects/issues/2, https://github.com/spinalcordtoolbox/detect-compression (I am also exploring several approaches by myself and trying to follow-up on [Etienne's work](https://github.com/spinalcordtoolbox/detect-compression), so far in the very very beginning stage -- code is not...
Maybe [PAM50/CHANGES.md](https://github.com/spinalcordtoolbox/PAM50/blob/master/CHANGES.md) might help:
> But, much to my surprise, the AP dimension wasn't cropped Interesting! An example for `-g 1` says that GUI does not allow to crop along the right-left dimension): https://github.com/spinalcordtoolbox/spinalcordtoolbox/blob/69c9125c4a6c497385c8cc6f56113421db5b633f/spinalcordtoolbox/scripts/sct_crop_image.py#L22-L23...
> The CSV is technically accurate Yes! > * How common is the 'S' orientation? I feel like in practice, I've mainly seen images in AIL, RPI, and LPI orientation....
Super interesting investigation! 💯 > This method essentially treats the binary mask's spinal cord area as equivalent to a rectangle, and the area as HxW -- we scale the sides...
Okay, I see! Thank you for the clarification!
Interestingly, when running `sct_compute_compression` without the latest changes proposed in https://github.com/spinalcordtoolbox/spinalcordtoolbox/pull/4176 using: ```console $ cd $SCT_DIR $ git checkout fd76016 ``` `sct_compute_compression` fails only for 7 subjects: ```console $ p118175@joplin:~/extrassd1/janvalosek/dcm-zurich_2023-08-09_persex/log$...
Notes from the SCT dev meeting on 2023-08-10: - Possible solution: output several individual files (one for each subject and each metric), then merge them - Check also `sct_process_segmentation`, which...
I also tested running the processing (see [here](https://github.com/sct-pipeline/dcm-metric-normalization/blob/d985ac0ca24d6253bfe717310f4c6a3529243b60/scripts/process_data_dcm-zurich.sh#L198-L206)) on `dcm-zurich` using `sct_run_batch -jobs 1` (i.e., not in parallel) and the processing finished without any errors. But it took 08h03m19s.
> When reviewing QC reports, in some cases we might want to add some comments Agree! This is an excellent idea! > Some ideas: > - new editable column >...