nyukat / GMIC

An interpretable classifier for high-resolution breast cancer screening images utilizing weakly supervised localization
https://doi.org/10.1016/j.media.2020.101908
GNU Affero General Public License v3.0
168 stars 48 forks source link

Mammogram scaling from 12- to 16-bit #34

Open AriBunnell opened 4 months ago

AriBunnell commented 4 months ago

Aloha! We are trying to run your model on our local dataset. We have mammogram DICOMs with BITSSTORED = 12 and BITSALLOCATED = 16. Can you please provide some more detail on the rescaling you performed to save your DICOMs as 16-bit PNGs? In another thread, you indicated that you scaled each image to 0 mean and unit SD; was this done before or after the scaling to 16 bits? A simple ordered list of preprocessing steps performed before cropping would be incredibly helpful. Thank you!