Add functionality similar to pydicom's skip_pixels

[felixhorger]

Hi,

I suggest to include functionality like in pydicom, where reading the pixeldata can be skipped.

The use-case I am concerned with:

• You want to get the pixel data of a single series from a data set obtained from a MRI scanning session.
• The pixel data is scattered over multiple single slice DICOM files all present in a single folder (... the pain is real)
• You could load all dicom files and then select the correct ones by uid, but depending on how large the dataset is, this might take ages
• Better would be to read everything but the pixel data, then select the right ones by uid, then load the pixel data afterwards of relevant files.

I have adjusted your code for doing that, let me know what you think.

Cheers, Felix

[notZaki]

Thanks for the suggestion @felixhorger ! I've ran into similar use cases too where reading the entire file is unnecessary---in my case, the scanner dumps all dicom files into a folder, and then I have a script to organize the files into subfolders that are named based on the series number and series description.

There is currently an undocumented feature to only read the dicom file up to a desired tag group. This can be specified by passing an optional keyword argument max_group. For example, dcm_parse(file; max_group = 0x0020) will parse everything including the 0x0020 group and then stop. Does that sound like a reasonable workaround?

[felixhorger]

That is indeed more elegant, thank you! :)

Cheers, Felix