Store measurements in dicom file

[przemekkr...@gmail.com]

I need to store some standard measurements (e.g. line, angle, cobb angle, etc.) in dicom file. My plan is to use the Presentation State but would be grateful for any advice on how to do this being standard compliant.

[David Clunie]

Presentation states are renderable on many PACS, but sacrifice the "semantics" of what you are storing.

I.e., it may look like a Cobb angle to a human, but to a machine its just a bunch of meaningless squiggles +/- disassociated strings.

Better is to use a DICOM Structured Report.

For guidance on encoding planar annotations in DICOM SR (including examples of Cobb angles) see:

http://dx.doi.org/10.13140/RG.2.2.34520.62725

[Kevin Archie]

On Tuesday, October 12, 2021 at 9:48:36 AM UTC-6, dcl...@dclunie.com wrote:
> Presentation states are renderable on many PACS, but sacrifice the "semantics" of what you are storing.

> ...

> Better is to use a DICOM Structured Report.
> For guidance on encoding planar annotations in DICOM SR (including examples of Cobb angles) see:
> http://dx.doi.org/10.13140/RG.2.2.34520.62725

I agree that SR is the right way to go in the long run, and even today when the use case is something quantitative (e.g., ML); but for now, if the intent is for humans to see the measurements, it's probably a good idea to also build a PS representation, which will display on top of the image on many viewers (I know Weasis and the Ambra viewer, but not Horos/OsiriX).

Unless maybe there's broader support for a visual interpretation of SR-encoded measurements than I'm aware? Would love to be wrong here.

- Kevin

[przemekkr...@gmail.com]

First of all, I would like to thank you for the link to the white paper. Also, I would like to ask if there is any SR in the IDC database that I could follow? Or maybe any other source?