PaLM Conf Minutes 2026-August-12

From IHE Wiki
Jump to navigation Jump to search

2026-08-12 IHE PaLM Notes

Attendees

Ralf Herzog, Gunter Haroske, Kevin Schap, Riki Merrick, Rob Rae, Bart Diricx (Barc), Richard Preston, Megumi Konda, David Clunie, Mustafa Qays Yousif, Ruben Fernandez, Kazunari Fukagawa, Patty Hernandez, Jim Harrison, Raj Dash, Scott Campbell

Agenda and Notes

CP Votes

  • CP-279, CP-281 and CP-283 passed – the comment has been resolved during ballot
  • Next Step:
    • Include these in revised DPIA document – David can update the document, needs the latest copy of the Word document (Kevin will share)
    • Alessandro to update the IGAMT profile accordingly
  • CP-275 – passed

DICOM Connectathon Update

  • Still ongoing till end of August
  • Annotations are supposed to be completed
  • Review of viewer behavior in the next 2 weeks (smaller number of viewers participating than who initially have signed up)
  • Compliance validation of images will have second round
  • Had good success with limited scanner and LIS (Meditech and Taiwanese University Group) participation of HL7 exchange
  • Results will be presented at Path Visions

DP Profile Updates

DPOW draft

  • Have the GitHub set up so Ruben will transfer the draft to the template and then send for review
  • Define a new workflow around quality control of WSI – how to prevent "bad slides" going to the scanner = DPQC

DPQC

This would be great to leverage DPIA and QC software output – Mustafa supports this project.

  • Do we also need a query/pull method to get QC information – similar to AIW-I in radiology?
  • Push workflow makes sense for the QC on scanning result
  • If the QC is part of the scanner, this is not needed for external interaction, but this would be helpful for downstream QC processes – this is where standardization is important
  • Result of QC and action based on decision – could trigger rescan, if reason for QC is due to scanner focus or something similar, which would need to include the parameter change needed to fix the QC issue could automate rescan vs re-cut of physical sample needed
  • We want to separate detection from decision
  • Out of focus (extend z-stack for field focus) and missed tissue (re-scan of full slide) – working with Hamamatsu on automating this; other scanner vendors not open to include automated QC follow up (on the edge of interoperability specs)
  • We can create standard protocols and procedure codes for the scanner and we want the way how to order this to be standardized
  • If scanner already has QC, then no need to send to third party QC tooling = triaging; might add failure notice into LAB-82 to let the human know first if QC is on the scanner (open issue #11 in DPIA) – provide error and location of scanner; then we can work on automation
  • Could we send the KOS without images to PACs?
  • If QC issue on WSI, then hold image forwarding for further evaluation will help reduce

Next Steps (DPQC)

Set this up as a project and phase this exercise across all issues already mentioned.

DPEC

  • Proposed to use Structured Data Capture (SDC) to create template for algorithm outputs
  • Need to see how to combine with DICOM annotations on the images (evidence for the result)
  • Communicating the result to the LIS needs support for some HL7 product-based communication (V2, SDC, CDA, FHIR)
  • Might have in-between artifacts
  • Should the synoptic report include hyperlinks to the DICOM annotations?
  • Some scanners add label on top of the overview, which can create issues for evidence creators (separate overview and label from the overview, to be able to more easily support de-identification)
  • Referencing reports to use SRs to link to SR – Ruben's statement – check transcription
  • Burned-in text is "evil" (but customers demand this) – send both?

PMW Update

In September or October. Goal is to have PRT aligned with the CP from Ralf.

AOB

  • Announcement: SHIELD F2F at HL7 WGM September 23 5:30 – 9:00 PM ET – please let Riki (rikimerrick@gmail.com) know, if you can join us

Next call: September 9, 2026

Chat Transcript

CP Votes / DP Profile Updates / DICOM Connectathon Update / PMW Update / AOB

Riki Merrick – Can you add the link to the spreadsheet here just for convenience?

Raj Dash, MD (Duke Pathology): YOU ARE VOTING VIA THIS LINK: https://docs.google.com/spreadsheets/d/1UbgL068W3JZ3y1KOqpQlITdP1unQt5UV/edit?usp=sharing&ouid=114293099465123885159&rtpof=true&sd=true

To submit your vote in the link above, use the "Comment" feature:

  1. Open the link
  2. Right click on the cell you want to comment on (Vote column/column F) and click Comment
    • Ctrl+Alt+M will also open the comment box
  3. Type your comment
  4. Click Comment

We have 8 in the final CP.

Raj Dash, MD (Duke Pathology): USE CASES: Detection of

  • Focus and clarity problems: Identifies out-of-focus fields, blur, and focal plane errors, triggering targeted rescans with higher-density focus points.
  • Specimen and mounting artifacts: Detects air bubbles, debris, coverslip issues, and broken coverslips.
  • Tissue and staining defects: Flags tissue folds, stain deposits, uneven staining, very faint tissue, and missing tissue areas.
  • Glass and labeling issues: Recognizes scratched glass, pen markings over tissue, and handles automated barcode and slide identification errors.

Ruben Fernandez (Sectra): AIW-I – It is currently Issue 11 on DPIA

Raj Dash, MD (Duke Pathology): For this meeting transcript: D. Clunie suggests separating DPEC results reporting from granular evidence creation. Standardized clinical reporting would benefit from using SDC. Granular measurements should use DICOM annotations.

Retrieved from "http://wiki.ihe.net/index.php?title=PaLM_Conf_Minutes_2026-August-12&oldid=136528"