4 ms·
I work on a PACS system. The problem isn’t DICOM, it’s big ass EHRs that don’t need to cooperate with anyone but themselves. Also, we’ve always had to allow ex
by batty_alex 4y ago
I work on a PACS system. The problem isn’t DICOM, it’s big ass EHRs that don’t need to cooperate with anyone but themselves.
Also, we’ve always had to allow exports in .iso/.zip with a viewer for the image data - going back to 2008. This is mostly forcing the big EHRs to stop playing games with patient data
Calling DICOM ancient is funny, the whole industry has basically been dragging their feet adopting it. FHIR will be “ancient” by the time it’s widely adopted, as well
- adolph 4y agoSome say ancient, others call it mature
- QuercusMax 4y agoLast time I seriously worked on interoperability between pacs systems in 2014, there were new products being put on the market from major vendors like GE that didn't support Unicode in dicom. And when I worked on radiation therapy systems, we ran into limitations with fixed-length fields being insufficient for the number of points in a region of interest, unless you use an encoding that doesn't indicate types. (4 bytes - either 2 for data type and 2 for length, or 4 for length with no data type listed). It's from the 80s and shows.