PaLM Conf Minutes 2023-December-06

From IHE Wiki
Jump to navigation Jump to search


Ralf Herzog Francesca Frexia
Kevin Schap Riki Merrick
Megumi Kondo Ruben Fernandes
Sam Spencer Gunter Haroske
Alessandro Sulis Kazunari Fukagawa
David Clunie Jan Schutrups
Dan Rutz Norman Zerbe
Hunter Putzke Mary Kennedy
Raj Dash Hynek Kruzik
Filip Migom

Next PaLM Meeting: January 10, 2024

Agenda Items

  • Re-vote on CP-LAB-267
    • CP267 – Motion to approve updated - against: 0, abstain: 0, in favor: unanimously – will check for quorum, if not achieved, then will proceed.
      • Will send to Mary Yungers – Jan 2024 so we can review the final version after that call – Riki will let Mary Yungers know about the plan.
  • ECDP meeting joint with DICOM WG26
    • DICOM WG26 has been working on a Connectathon prior to the conference.
    • DICOM WG 26 has not discussed meeting jointly with PaLM.
    • If we are close to finishing DPIO and DPIA updates after Porto we may want to meet with WG26 at ECDP.
      • World like to start on the evidence creation.
      • The DICOM WG26 Connectathon is annotation focused.
      • Vendors should be at ECDP
    • Pre-conference will be available on June 4 and 5th – no problem for rooms for that time per Norman. Just let him know.
  • Digital Pathology
    • Examples for SPM-6 will be ready for January meeting.
    • Questions from Jan
    • IHE PaLM Image 1 12-6-2023.png
      • LAB-80 message
        • Unsolicited or as response to LAB-81
        • LAB-82 as output
      • LAB-80 questions:
        • Where would the studyID go?
          • Should be created by the manager and the scanner should use them
        • Accession ID – is this really O?
          • No, it is M.
          • Datatype includes the assigning authority = DICOM issuer
        • Reference Request Sequence = code for the procedure codified in OBR-4
          • In DICOM we allow breaking down of the steps.
          • In V2 we would be using OBX segments after OBR to describe specific parameters for the scanner.
          • We could come up with a list of universal protocols.
          • For implementations we map the scanner equipment procedure codes to the LIS procedure codes to understand the process steps.
        • Variation on specimen processing is communicated in OBX segments after SPM.
        • DICOM design supports making images being able to be detached from the AP list access
          • Include all the details around the processing
        • Can create resolutions by using pixels or with the lenses in the scanner – how to deal with that?
          • In DICOM, these are pixel spacing attribute for any particular layer, defining the resolution (if it was down-sampled or up-sampled – has also field called objective lens power – so can be recorded in DICOM).
        • DICOM followed the HL7 Specimen DAM
          • Fixation, staining,
          • What the specimen actually is
            • DICOM encourages use of codes for body site = SPM-8
      • LAB-82
        • Derived / primary / resampled – where does that go?
          • DICOM object
          • Vendors differ in how they encode those in different patterns.
          • It should be discussed with WG 26, not a PlaM issue (unless we need to represent this in DPIA.)
          • How do you know you are not missing image layers?
            • If it is not there, it does not exist.
        • Scan power
          • Objective lens power is supposed to represent the lens power with which the image was made.
    • Goal is to have all the information in the messaging so we can fully automate the
      • Do we need to consider using an equipment command message?
  • Proficiency testing profile updates and review of mappings - to be discussed at January 10, 2024 call

From Chat

08:24:01 From Raj Dash to Everyone: So Scanner does NOT create

08:25:05 From Gunter Haroske to Everyone: The Specimen Identifier in LAB-80 is not the BlockID but the Slide(better Section)ID?

08:26:22 From Rúben Fernandes to Everyone: Replying to "Checking my notes..." DPIA does not define that I believe. But for the sake of generating required elements, the scanner may generate if not available.

08:31:17 From Gunter Haroske to Everyone: Replying to "Checking my notes..." It should be available by the LIS which produced it.

08:36:35 From Raj Dash to Everyone: We probably need to align DPIA to accommodate pixel resolution (dpi) along with optical magnification.

08:37:21 From Raj Dash to Everyone: It is important for comparing images should populate the image file

08:37:37 From Gunter Haroske to Everyone: This would be more useful than magnification alone

08:37:56 From Raj Dash to Everyone: But that is up to the scanner... so not sure we need to accommodate in DPIA UNLESS this is configurable and needs to be part of the order

08:38:23 From Raj Dash to Everyone: I don't think I can adjust DPI

08:38:30 From Raj Dash to Everyone: Only optical magnification

08:38:41 From Raj Dash to Everyone: The DPI resolution is fixed for every scanner, I think...

08:38:54 From Raj Dash to Everyone: Because imaging sensor fixed

08:40:25 From Raj Dash to Everyone: I will make sure to accommodate those Jan in the examples

08:42:11 From Raj Dash to Everyone: If we specify, then LIS will start using. I don't think there is a technical limitation

08:43:05 From Dan Rutz to Everyone: Agreed there shouldn't be a technical limitation there.

08:47:49 From Raj Dash to Everyone: Why not store the ratio ? What is the downside?

08:50:46 From Rúben Fernandes to Everyone: It can be inferred, so having it there wouldn't add a lot to a viewer

09:00:54 From Kevin Schap to Everyone: Reach out to me for issues you would like to have added to DICOM WG-26 agenda