Re-use Display-Set Images

When creating display sets for a reader study, it is sometimes the case that the same image is being used multiple times in different display sets. There is no need to upload the image for each display set. Once uploaded, the image reference can be used when creating additional display sets. This reduces storage usage and optimizes performance.

Speed Benefit

Viewer applications such as CIRRUS need to download each image in the display set. If a display set correctly re-uses an image, CIRRUS can re-use a previous download. This generally speeds up the case loading.

Get an image reference

Archives, Job results and other display sets have references to already uploaded images. If you have not yet uploaded the image, you can first create a display set.

Create the first display set

Create a display set roughly as follows:

reader_study_slug = "my-reader-study-slug"

case = {
    "ct-image": "/path/to/image/file/00.mhd"
}

# Create a simple display set with one MHD image
display_sets = client.add_cases_to_reader_study(
    reader_study=reader_study_slug,
    values=[case]
)

Wait for processing

The postprocessing of images on Grand Challenge means that we have to wait a short while until all files are processed:

import time

reader_study = client.reader_studies.detail(slug=slug)

while True:
    # Fetch the latest status of all display sets
    display_sets = client.reader_studies.display_sets.iterate_all(params={"reader_study": reader_study.pk})

    # Check if all display sets have their values processed
    if all(len(ds.values) > 0 for ds in display_sets):
        break

    # Wait a bit before checking again
    time.sleep(60)

Create a new display set

Finally, when the images have been processed you can reference them when creating a new display set:

image_reference = None

# Reference the first display set, and the first value
for value in display_sets[0].values:
    if value.interface.slug == "ct-image":
        image_reference = display_sets[0].values[0].image

assert image_reference

new_case = {
    "ct-image": image_reference,
}
new_display_sets = client.add_cases_to_reader_study(
    reader_study=reader_study_slug,
    values=[new_case]
)