[carbocation]

I use this and it's great!

One thing that would be helpful (and perhaps it's already included and I just haven't seen it?) is a set of functions to make the parsed image appropriate for display.

Right now, the raw pixel values are given. To turn this into an image, you have to determine the word size (e.g., 16 bits), then interpret the results in the context of the window and center values from the metadata. For my current batch of images, for example, unless I do postprocessing, the images just look black.

[suyashkumar]

Thanks! Indeed, right now if you're opening a DICOM with Native PixelData the API gives you the raw PixelData without manipulating (normalizing) it. This is so that you can apply WindowWith and WindowCenter settings to your choosing, should you want to (and sometimes, there are multiple window withs and centers to choose from in the metadata).

I recently introduced a CommonFrame interface that wraps both Native pixel data and encapsulated pixel data that gives you a clean Go stdlib image.Image to use for post processing (no matter the underlying data). Introducing some options to this that allows setting of window width and level would also make sense!

[carbocation]

Exciting! I'll have to take a look at the new interface. Thanks!