Quantitative imaging and Radiomics

Clinical CKM publication
1

Dear all,

We are currently working on using openEHR for quantitative imaging and radiomics in the HiGHmed project. Radiomics is the extraction of quantitative information from images. For those of you who are
not familiar with radiomics I recommend these papers that explain the concept of radiomics very well: https://doi.org/10.2967/jnumed.118.222893 and https://doi.org/10.1016/j.ejca.2011.11.036
There is also a video that gives a good introduction: Intro to Radiomics_30min_Lambin_Nov20 (1) on Vimeo

Radiomics involve information about the original images, the derived images and the radiomics measurement method and features. Our aim is to store information about the radiomics measurement method and the radiomics features as well as metadata about the segmentations in openEHR. Therefore, I already designed some archetypes that can be used for the data from our radiomics pipeline. For the radiomics measurement and feature I created this archetype: https://ckm.highmed.org/ckm/archetypes/1246.145.1281
For segmentations I created this archetype: https://ckm.highmed.org/ckm/archetypes/1246.145.1279
My thought was that these archetypes extend the archetype openEHR-EHR-OBSERVATION.imaging_exam_result.v0 in the slot ā€œimaging findingsā€ like I did in in this preliminary template for a radiomics report: https://ckm.highmed.org/ckm/templates/1246.169.1140

I am currently working on the revision of these archetypes. As we would like to contribute our archetypes to the international community, I would like to ask you about your opinion on the archetypes and how I intended their use. In particular I have the following questions:

  • How can derived imaging like segmentations be represented in openEHR?
  • Can derived images and radiomics features also be considered as an OBSERVATION in openEHR?
  • Is it correct to extend the archetype openEHR-EHR-OBSERVATION.imaging_exam_result.v0 for derived imaging and radiomics information or would you consider this not as an imaging examination result?

Best regards,
Christian

4 Likes
2

Hi there,

just a friendly reminder :slight_smile: . Is anyone out there who is involved both in openEHR and quantitative imaging?

Best regards,
Christian

1 Like
3

Hi Christian,

This is well outside my field of clinical expertise but the overall approach looks pretty good to me

extend the archetype openEHR-EHR-OBSERVATION.imaging_exam_result.v0 in the slot ā€œimaging findingsā€ like I did in in this preliminary template for a radiomics report: https://ckm.highmed.org/ckm/templates/1246.169.1140

  • How can derived imaging like segmentations be represented in openEHR?

Your approach seems sensible to me

  • Can derived images and radiomics features also be considered as an OBSERVATION in openEHR?

Yes - perfectly acceptable, even if they re derived/calculated features, to me they are still Observations.

  • Is it correct to extend the archetype openEHR-EHR-OBSERVATION.imaging_exam_result.v0 for derived imaging and radiomics information or would you consider this not as an imaging examination result?

I agree with this approach