Tiger challenge WSI data variation

It was mentioned that: "Note that the manual annotations released with TIGER refer to 0.5 um/px magnification and may not be directly applicable to some of the original TCGA-BRCA slides (originally scanned at 0.25 um/px)." Request to clarify the following points regarding this:

1. Could you please point out some example tiff images (training, validation/test) in the TIGER datasets which has 0.25 um/px.
2. Could you also suggest if we can use any meta information (of the TIFF headers) to differentiate between 0.5 and 0.25 um/px data? So far we observe, there is a parameter named TiffResX/TiffResY, with their values ranging between 15K to 23K (approximately). Its reciprocal comes close to 0.5um/px. However, we could not find any data corresponding to 0.25um/px (i.e. TiffX/YRes ~ 40K).
3. So far as we saw, the scaling factor between the successive layers of the TIFF pyramid was always 2, but the number of layers varied from 5 to 9 for different images. Under the circumstances, should we assume that layer-0 always corresponds to 20x magnification and the lowest layer varies from 0.6x to 0.04x?

Dear Vaibhav Singh

1. We converted all slides to have ~0.5 spacing as the lowest layer 0, and therefore we can not point you to an example with ~0.25 spacing

2. You can use meta information to get spacing information. For example, with the wholeslidedata package, you can get the exact spacing from an opened wsi with wsi. get_real_spacing(0.5). Please note that there are no ~0.25 spacings in any image.

3. All slides used in the challenge will have a ~0.5 spacing at level 0. And the subsequent levels are always downsampled by a factor of 2. However, there is no guarantee that a specific level will be present in a slide. Small slides tend to have fewer levels. However, If your code relies on a missing level, you could downsample it yourself.