Closed surajpaib closed 8 months ago
Hmm, I'm not too sure how you'd want the format to be for these on the release (since we don't have any DICOM + json workflows yet and I am not too sure how to do that myself without spending a lot of time 😓 )
So I've attached both the features.json and the centroids.json for the IDC DICOM scan that I linked above. Maybe you have an idea of how to do this best?
That's great! I'll implement the dicom workflow later today!
@LennyN95 Here is another sample.
Image: s3://idc-open-data-cr/18b83380-0f27-42f2-af7c-65e2d23668ac/*
Centroid:
LUNG1-076_centroids.json
Features:
LUNG1-076_features.json
/test
sample:
idc_version: 17.0
data:
- SeriesInstanceUID: 1.3.6.1.4.1.32722.99.99.298991776521342375010861296712563382046
aws_url: s3://idc-open-data-cr/40db0070-9773-413a-8d7e-c91c37c0006b/*
path: dicom
- url: https://github.com/MHubAI/models/files/14520712/centroids.json
path: 1.3.6.1.4.1.32722.99.99.298991776521342375010861296712563382046.json
- SeriesInstanceUID: 1.3.6.1.4.1.32722.99.99.309695689942006699082558764031786785731
aws_url: s3://idc-open-data-cr/18b83380-0f27-42f2-af7c-65e2d23668ac/*
path: dicom
- url: https://github.com/MHubAI/models/files/14528031/LUNG1-076_centroids.json
path: 1.3.6.1.4.1.32722.99.99.309695689942006699082558764031786785731.json
reference:
url: https://github.com/MHubAI/models/files/14612917/fmcib_mhub_default_output.zip
I tested the default workflow, made some small fixes and it all looks good.
Also tested the slicer workflow on a folder structure that looks like this
Using the
PatientID
folder as the lead for the structure. It all works very well 🎆For the testing, This should pull all the DICOM data:
s3://idc-open-data-cr/40db0070-9773-413a-8d7e-c91c37c0006b/*
and I'll make a release with the centroids.json and the features.json