Closed cooperlab closed 5 months ago
This could be done by having a struct ruleset with a fallback rule for unknown tags/images which removes those tags/images. We'd have to add a fallback rule.
A good test here is can we read the images with large_image or openslide, and can they still be opened in QuPath.
I've been doing some reading of the tiff standard to try and figure out how to determine what the minimum tag set would be. A good starting point would be the baseline tiff plus any tiff extensions commonly used by whole slide images. I'm not sure what those would be, however.
I imagine that the "Tiled Images" tag set should definitely be included, but the others are unclear.
I believe this screenshot captures all of the tiff extensions defined by Tiff 6.0. Which ones should be included in the minimum tagset for this sort of redaction?
The required fields are plus 254, 266, 274, 317, 320, 322, 323, 324, 325, 330, 338, 339, 346, 347, 512-531, 559, 34675
Define the minimum required metadata necessary to load .svs. This will allow users to appy a "strict" redaction profile without requiring the institution to inspect a bunch of tags. Start with .svs since the viewers and dataloaders are more mature.