Rad Tech Minutes 2020-04-27-28

• AI Results checklist
  • Did we line-by-line the entire document
    • Not yet. Scheduling additional calls for page-by-page review
  • How ready is it to go out for TI: Completely, Almost, Soonish, Hmmm
    • Soonish to Almost - editor still has a fair number of todos, and there are some discusses left
    • Calls have been scheduled
  • How did the work fit in the allocated bandwidth? (Time to spare? Just right? Things were left undone?)
    • Overflowed allocation - Likely a record number of comments (464 vs 80-150)
    • Some points of controversy - mostly they were anticipated -
    • Some comments were people new to IHE Rad and answers
    • [Editor wants to evaluate the balance of questions/topic/type etc]
  • Review the evaluation. Which complexity/uncertainty/effort points missed the mark?
    • [Editor will try and tally offline]
    • Filtering was flagged but used a bit more UP
  • Or alternatively, estimate how many points you went over and assign the overage effort/complexity/uncertainty to the appropriate points.
    • [Editor may try depending on available time]
  • Are all the open issues closed?
    • Yes. All 24 of them.
  • What significant debates in TI-prep were not anticipated in the Kickoff or PC-Prep
    • (broadly more details to work through than expected)
    • Variations on TID1500/1410/1411 - thought they were "more well defined" in DICOM but we found a lot of variations and questions around different results - was not expected.
    • A few of the topics were "cans of worms" - simple on the surface but loads of complexity lurking
  • Did the Breakdown of Tasks accurately reflect the work? What extra tasks arose?
    • Shape was mostly correct. Just more work/details
  • What residual risks are worth noting
    • Have bumped a couple things to "future work"
    • Potential complexity for Displays might be high
    • Existing risk of getting "clean text" for TI on schedule
    • Potential for complex results (risk existed before but now we partly own the problem)
  • Does it feel we've met all the use cases
    • Yes.
    • Primitives seem to cover well.
    • Requests to cover FHIR/EMR (hits scope control)
  • Did the promised resources manifest
    • Yes. Way more reviewers than expected.
    • Would have been nice to have more Evidence Creators joining the call live.
  • What vendors are engaged (for each actor)
    • Evidence Creator - "conventional radiology" vendors, but maybe not newcomers as much
    • Image Manager - usual suspects (GE, Siemens, Philips, Canon/Vital, Change, Visage, IBM, ...)
    • Image Display - usual suspects (GE, Siemens, Philips, Canon/Vital, Change, Visage, IBM, ...)
  • Who should specifically be targeted for TI notification (implementors & advocates)
    • Teri - RSNA Demo & Mailing List; ACR DSI; DICOM WG-23; SIIM Hackathon;
  • When will we have sample data/objects
    • Great question.
    • Limit profile appendix to showing what spec looks like in an instance, but not to show the various things you can do.
    • The latter goes in the test objects archive and we need to recruit for that, perhaps with Teri (RSNA Demo), Lynn (Connectathon), etc. And Steve's Shareathon concept.
  • Was the profile where it needed to be at the start of the TI meeting, if not what was the gap
    • Not completely, Would have been good to have more TODO's done and fewer Discussion's open (more could have been Review if editor had more time to prepare proposed resolution), but with the volume of comments, that was a bit unavoidable.
  • Was the profile where it needed to be at the end of the TI meeting, if not what was the gap
    • No.
    • Gap = too much work to cover. Got 310 comments done, but still have a double handful to discuss, and about 100 todos for the editor
    • Approach to comments (do offline review/flagging to discuss) helped get through the number that we did.
      • Editor will try to write up approach and tallies
  • Do you need any tcons between now and TI Publication
    • Oh yeah.
    • Scheduled one more discussion resolution call, and three more review calls (discuss review items and do page-by-page).

• AI Workflow for Imaging checklist
  • Did we line-by-line the entire document
    • No. Not yet. Call is scheduled for page-by-page
  • How ready is it to go out for TI: Completely, Almost, Soonish, Hmmm
    • Soonish to Almost - lots of Todos - Discussion points remain and will work through them too
  • How did the work fit in the allocated bandwidth? (Time to spare? Just right? Things were left undone?)
    • Overflowed time budget.
    • Spent lots of time re-debating UPS-RS
    • Didn't get to worklist priority
  • Review the evaluation. Which complexity/uncertainty/effort points missed the mark?
    • Use Cases underestimated CP and EP
    • Workitem Mech - chose UPS and didn't allocate time to debate UPS-RS but did
    • Request Model Inference & Push Pull - underestimated UP and CP - Push/Pull discussions were extensive,
    • Priority Notification - underestimated UP and CP
    • Performer Exploring for Inputs - seem to have moved away from that so effort mostly OK?
  • Or alternatively, estimate how many points you went over and assign the overage effort/complexity/uncertainty to the appropriate points.
  • Are all the open issues closed?
  • What significant debates in TI-prep were not anticipated in the Kickoff or PC-Prep
  • Did the Breakdown of Tasks accurately reflect the work? What extra tasks arose?
  • What residual risks are worth noting
  • Does it feel we've met all the use cases
  • Did the promised resources manifest
  • What vendors are engaged (for each actor)
  • Who should specifically be targeted for TI notification (implementors & advocates)
  • When will we have sample data/objects
  • Was the profile where it needed to be at the start of the TI meeting, if not what was the gap
  • Was the profile where it needed to be at the end of the TI meeting, if not what was the gap
  • Do you need any tcons between now and TI Publication