search

marksgraham / OCT-Converter

Tools for extracting the raw optical coherence tomography (OCT) and fundus data from proprietary file formats.
https://pypi.org/project/oct-converter/
MIT License
193 stars 70 forks source link

convert to nii.gz after extraction #100

Closed benjaminralex closed 10 months ago

benjaminralex commented 1 year ago

Hello,

Thanks for the awesome work! This has been a really helpful tool.

After extracting the necessary volumes as numpy arrays, I would like to store them as nifit files. For that, I would need the necessary pixel and slice spacing values. Is this something that can be extracted from the original files?

Have you encountered this when displaying or visualizing the extracted arrays?

Thanks.

marksgraham commented 1 year ago

Hi,

@naterichman has just begun work on a feature allowing you to write to dicom, which will also need pixel and slice spacing, and a .nii saver could piggyback off a lot of these features. But it might take a while before it is available.

Currently we are able to read more data than we actually store to the OCT/Fundus classes. You can see what is extracted for each reader here. I've been thinking of implementing a .get_all_metadata() method in the readers that will get all the metadata we're able to extract from a file and store it as a dict in the class, so that users are able to do what they like with this data - you should be able to do this to write your own .nii reader, while we wait on a more complete solution that mimics the dicom writer. Does that sound useful? If so, is there a particular filetype you're interested in? I could start implementing for that filetype and build from there.

benjaminralex commented 1 year ago

That would be perfect actually!

E2E, ZeissDCM, and FDS would be great starting points. Really all of them, because downstream viewing and packaging into dicom/nifits is not possible without the metadata.

I might try to create a get_all_metadata method for my specific application. If successful, I will definitely pass it along.

Thanks a lot. Much appreciated 👍

marksgraham commented 1 year ago

OK sounds good, let me know if you need any help!

marksgraham commented 1 year ago

Hey,

Just flagging that I've added a .read_all_metadata function to fda and will add to fds soon, here: https://github.com/marksgraham/OCT-Converter/blob/ebd22e560779eb442b4666451244bbccf28f0958/oct_converter/readers/fda.py#L217

marksgraham commented 10 months ago

There is now functionality to export to dicom files, too. Closing this for now.