PaLM Conf Minutes 2019-May-27-29

From IHE Wiki
Jump to: navigation, search

Attendees

27-May 28-May 29-May
Raj Dash x x x
Francois Macary x x x
Riki Merrick x x x
Nobuyuki Chiba x x x
Francesca Frexia x
Yoshitake Fujisaku x x
Takuya Haga x
Satohiro Hamano x x x
Gunter Haroke x x
Jim Harrison x x
Ralf Herzog x x x
Yoshimi Hirasawa x
Naomi Ishii x x
Chihaya Kakinuma x
Kenichi Kamijo x
Tsuyoshi Kawata x
Mary Kennedy x x x
Hiroyuki Kohda x
Megumi Kondo x x x
Filip Migom x x x
Ichiro Mori x
Yosuke Murai x
Takeshi Nagaya x
Yuichi Naraki x
Manabu Nomura x x
Maki Ogura x
Yasunari Shiokawa x
Nozomi Sugiura x
Alessandro Sulis x x x
Akitoshi Suzuki x x x
Kenichi Takahashi x x x
Haga Takuya x
Kiyoko Tateishi x x
Ikuo Tofukuji x
Fumio Yashiro x
Eiji Yoshimura x


Monday, May 27, 2019

Introductions were given and Francois reviewed the agenda

  • Hiroyuki Kohda from Hitech is the JAHIS member
  • Akitoshi Suzuki is the IHE Japan Pathology Chair
  • Yoshitake Fujisaku is the IHE Japan Lab Committee Chair

Francois reviewed the assets publication timeline:

  • LCC – public comment resolution is in process
  • Digital Pathology
    • white paper – initial draft was shared beginning of this year, working through comments
      • will review during this meeting
      • should be ready to publish by June 3
      • Will use acknowledgement section to capture all the authors
      • can we also submit to journal to get more vendor input for future profiles:
        • Journal of Pathology Informatics
        • Archives
        • JAMIA
    • Image Acquisition Profile (DPIA)
      • Several open issues for review at this meeting
      • Joint calls with DICOM WG26
      • We can publish with open issues, if needed
      • For Public Comment June 15 (priority over the white paper)
      • For TI – mid September
    • Ordering Profile (DPOR)
      • Need to start work/ Francois and Raj – draft by end of June for IHE PaLM review
      • Can reuse APW profile work / LAB-1 and LAB-3 update
      • For Public Comment: July 17
  • SET = Specimen Event Tracking
    • Vol 1 is written
    • Vol2 is drafted, but need new message types => HL7 Change Request to finalize and submit to HL7 OO June 4 – also Harmonization proposal due by June 24 / harmonization call is July 17
    • For Public Comment September
  • For TI: November - for Europe and Japan Connectathon 2020
  • Transfusion Medicine
    • TMA was in Public Comment April 2018
      • TI: November 2019
      • Filip has 3 system vendors interested- and customers voiced interest in implementing the interfaces with EPIC
      • For European Connectathon in 2020 (is in Brussels)
  • LSH
    • On hold until we can find new champion
  • PaLM TF:
    • CP inclusion
  • Publishing Priorities:
    • For US Connectathon
      • DP IA
      • DP White Paper
      • LCC
      • DP OR
      • PaLM TF v10
    • For Europe Connectathon
      • TMA
      • SET

Digital Pathology - IHE PaLM Digital Pathology Outline and Analytical Exchange on Digital Specimens PowerPoint

  • White paper:
    • Acknowledgement section – Co-Chairs will work out the format
      • Organization
      • Individuals
    • Table of Contents needs to get fixed (Francois -> done)
    • Check the template format for IHE white papers (Francois/Raj)
    • Remove the details of the transaction numbers in the journal submission
    • Include the citation for HL7 Specimen DAM April 2019 (Riki to send to Raj)
    • Add an image to explain the planned IHE profiles
    • Ordering profile = DPOR
        • Does this cover the order of the tissue exam or is it for the image ordering?
          • We may need both
            • Need convergence of LAB-1 and PAT-1 / LAB-3 and PAT-3 so that can support both the clinical, the micro and the pathology lab order
            • Identify the slide that needs to be turned into an image in the order
            • Open issue – what to do with LAB-2 for documenting reflex orders
              • Order group number (ORC-4) can be used to group all the reflex test together, so we no longer need LAB-2. Issue closed.
            • There is both test-based ordering or specimen based ordering in AP
    • Image Acquisition = DPIA
    • Evidence Creation = DPEC
    • Image Ordering = DPIO = this is similar to LAB-4/LAB-5
    • For the reference to Specimen DAM, keep Appendix mapping and only include a few sentences on the effort; do not need the detail on the elements in the text
    • Raj will share the updated version via the google group
  • DPIA:
    • Diagram reviewed
    • IHE Glossary – Riki to check with Heather Grain to see if all of those were added to SKMTglossary.org site – if so, will need to add these there also.
    • OBX segments will come back – still need to define the LOINC codes to use for each of these
    • Some open items will be moved from the spreadsheet to the document open issue section (Raj)
      • Error codes from the scanner
    • For query, we have query response come back before LAB-80 – the RSP message can only report on errors on the query itself, no more. The work that needs to be performed is sent in LAB-80 for the IWOS: it can be a NW or a CA for the specific slide: ORC-1 = DC; if we need to communicate a reason for the fact, that there is no IWOS for the sample ID sent – ORC-16 (is optional in IHE LAW) using HL70949 codes – but why would it make a difference to the scanner why there is no IWOS?
    • If the modality cannot perform the IWOS sent in LAB-80 – the modality shall send back the application response of LAB-80 (example request to scan @ 40x magnification, but the scanner is @20x only) ORC-1 = UA – see section ID in LAW = TF 9.0 Vol 2b - 3.28.4 transaction diagram (Riki to send Word doc to Raj)
    • Open issue:
      • Issue 4: yes - the modality sends observations back – for example the thumbnails, as well as some other IDs
        • Also, error messages per LAW can be in OBX-5 or OBX-8 (MAKE a comment
      • Issue 5: Use the same HL7 version as LAW = v2.5.1
      • Issue 8: Image may be stored elsewhere before these images are stored in the archive
        • thumbnail images can be viewed and edited, deleted before they are permanently stored
        • Does that mean LAB-82 will not come back until the image has been stored in the archive – temporary local storage will not be sent back?
        • Do we want to support OBX-11 = P (preliminary)? Yes
        • When OBX-11 = F, the understanding is that this is used only once the image is stored permanently?
        • Image manager is not defined as the final PACS in our profile, so in the described issue you would collapse the image manager, the acquisition manager and image archive into one system
        • OBR-49 can convey the information that the sent link to the image is not persisted
        • Are there Japanese scanner vendors interested? Possibly Hamamatsu, Olympus– so make the support for RAD-9 and RAD-10 optional for the acquisition manager and then need to add a section describing an option so scanner vendors can declare
      • Issue 9: Glass slide ID for multiple orders – do we need to create an assumption that the acquisition manager MUST support sending the orders with the proper timing – OR require the profile to make changes to the ID on the physical asset? Per Raj, currently adding the stain name onto the slide – so we could combine the slide ID with the stain name to make it unique and use that as the slide ID number for the IWOS – rephrase that first sentence to make clear that the issue occurs, when the same glass slide has different physical properties, each with different IWOS
        • Remove the option of adding the IWOS ID onto the slide – this is hard to because of on limited space
        • IWOS IDs should be purged ONCE it is completed – will that work? Can we remove this one? yes
        • Or, do we recommend assigning a new ID after each processing step = e.g. a version number

SET - SET Profile and HL7 CP Update PowerPoint and HL7 CR SET with Comments

  • Francois supports the use of different trigger events for each business trigger because it matches the ADT paradigm as well as still leaving the EVN-4 open to allow it to send reason for the event. For example, the trigger event of missed specimen collection could then be sent in EVN-4, which is what this field is really meant for.
  • In the SSU message we use the same message structure for the same kind of thing – so any specimen movement is the same kind of concept.
  • We will use the different trigger event codes - one for each business case (Alessandro to document this decision in the CR)
  • Event Performer – change to Event Participant and change segment to segment(s)
  • Need unique Event ID – ADD NEW Field to EVN segment – if the event is being re-sent, the MSH-10 would change, but the ID of the event would not change- -hence the change
  • Change the requirement for EVN-4 from SHALL to MAY since this is the base standard – will decide if SHALL or SHOULD in the IHE profile later - we could make a few more suggestions for the user defined table in the base => harmonization proposal
  • We may be able to reduce numbers of message structures further – at the end of the use cases
  • Should we add OBX after each of the SPM and SAC segments – Add an optional CONTAINER group and optional multiple OBX after SPM. (Riki will fix)
  • For specimen collection do we need all the detail about the collection event itself, or do we want to just note, that the event happened?
  • For specimen movement MUST have PRT – so remove the []
  • The SHP segment was removed because it did not fit as a group with the first element optional, it also does not work as segment under SPM
  • Riki to add the optional OBX segments for SPM and SAC. This applies to ALL of these message structures
  • For From Entity = change to Event Participant = From Entity = PRT-4 will be identified as code per NEW Harmonization proposal for HL70912
  • EVN-4 is now still available for the reason of rejection – do we still want to use SPM-21 / SAC-8?
  • Can you send message that you accept the specimen and then reject one of the many containers? MUST have at least ONE PRT
  • Package centric:
    • PRT also should not be optional = take off []
    • Looking at the shipment message – SHP carries similar information as what we cover in the EVN plus PRT, except confidentiality and number of packages

PaLM Activity Around the World - Lab and AP Update Spring 2019

  • Japan
    • There is no organization change in IHE- Japan and JAHIS
    • An IHE update was given at the 38th Medical Informatics Joint Convention in November 2018
    • JAHIS conducted an infection control course for LIS vendors in March 2019
    • JAHIS continuously participates in the standardization committee to clarify/improve issues in order that JLAC11 is widely used.
    • The Digital Pathology Technical Committee report: IHE Japan applied a proposal to update standard as version 3 based on revised Medical Treatment Fee 2018; adding guidelines by the relevant academic societies.
    • There was an IHE-Japan and JAHIS joint meeting in March 2019.
    • IHE-J Connectathon will be October 7-11, 2019. Profiles to be tested include: LTW, LTW-MB, LDA, LAW, LPOCT, and LBL
  • US:
    • LAW has now been published as CLSI Standard AUTO-16 – otherwise no update
    • US 2019: Connectathon only 2 actors tested XD-Lab
  • Europe: no update

TMA - Blood Transfusion Administration Workflow PowerPoint

  • Filip reported that several parties in the Netherlands and Belgium are interested in this profile. MIPS and Epic are ready to implement, so looking for final TI profile as soon as possible
  • Ordering profile – not currently defined, so MIPS is using LAB-1, with a few additional restrictions
  • Ordering use cases:
    • How long will it take you to find bloodtype X?
    • If the request is for 3 blood bags and then add on 1 extra blood bag, but in the message, they may send it as an order for 4 blood bags (without canceling the first order of 3, because that would possibly stop the whole process, including the crossmatching etc.)
  • Dispense profile - not currently defined, so MIPS is using LAB-3 with a few additional restrictions – a simplified OBX segment(s) is used to convey the status
  • Administration:
    • How will a pooled group of plasma identified on the blood bags be filled in – FDA just approved this, not clear how the labeling works – seems that the flag would be different for each bag in this case (00 – 99) – is there also a serial number
    • ISBT218 is used for labeling formatting in Europe – also in US. Germany invented a different standard, but is still closely related
    • Donor identification (BTX-2 related) is part of the uniqueID – PLUS Product Code?
    • Should “suspect” be “suspend”?
    • Could we use SET to cover the requirement to track movement from the blood bank to the ward – maybe covering by specimen movement = but this is a product, not necessarily a specimen.
    • Make Use Case #1 the “happy path”
    • Make use case #2 the interrupted occurrence (interrupt and end) – this could have 2 versions (interrupted, restarted and end).
    • Use case 3 – adverse event AFTER completed transfusion
    • Use cases 1 and 2 use LAB-70; use case 3 uses LAB-70 PLUS another message for the adverse event
    • A positive patient ID is what is important for the clinician
    • The profile currently does not cover erroring message. If the blood product ID does not match the Patient ID per the cleared list of blood products; BTX-2 PLUS BTX-3 together gives unique ID – but what should be used for unique ID = Barcode or human readable content of the barcode?
    • In UDI this is the required element: “The full UDI carrier of the Automatic Identification and Data Capture (AIDC) technology representation of the barcode string as printed on the packaging of the device - e.g. a barcode or RFID.” It also supports a human readable version: The full UDI carrier as the human readable form (HRF) representation of the barcode string as printed on the packaging of the device
    • BTX-11 = status – uses HL7 defined table HL70513 – missing begin and interrupted
    • BTX-14 and BTX-15 for performer and verifier (make this R when BTX = code for begin) – neither repeats per the base;
    • In the workflow MUST have the performer and who administers and who authorizes (that would be ORC-12) – only performer will change for each of the status changes
    • Adverse reaction BTX-18 and BTX-19

IHE PaLM 2019 webinar:

  • Riki and Raj will give it
  • The focus will be on digital pathology
  • It will also talk about the work of TMA and SET
  • There will be a prep call in July

Tuesday, May 28, 2019

Review of Laboratory Clinician Communication Profile Memorandum

LCC:

  • A comment in Google spreadsheet: When order filler is aware of insurance coverage of patient to make the more equivalent test that could be covered. Will add as an example to the scenario for incorrect order
  • CAP Quality Practices Committee comments: (See link above)
  • Comment regarding asking for arbitrary clinical communication: the LCC is ONLY in the context of the orders; additional questions on orders to get more information to decide on the right testing. LCC can send back as ask at order entry questions (AOE) in return, but to ask the questions would need more work in future (not sure how the questions without the answer would come across (OBX / NTE). Add this comment as an open issue so we can investigate and decide if we will adjust prior to final text or create a new profile
  • AOEs are usually well defined – they seem to ask for more random questions about the existing order, but without the recommendation component need a different message structure.
  • Recommended orders for specific situations in CPOE systems – that is more clinical decision support on the CPOE side now, not coming from the lab, as in this profile
  • LCC must support AP – pathologist should be able to make the recommendation for the new test – LCC does support that (we don’t have restrictions on where the order comes from PRIOR to the message being created).
  • LCC does make a statement that this is applicable to all order situations; but we are focused here on specimen related orders. We would have to derive a profile for that, like dietary orders. The base standard was updated to support these transactions, so other domains can use the same approach
  • Jim had submitted a few comments, that didn’t make it into the spreadsheet – based on comments from UVA colleagues:
    • They had difficulty that LAB-6 and LAB-7 are not needed to be used together – but we have 2 separate use cases – recommend in Vol 1 to ensure those use cases translate into the separate transactions. (Jim will update)
  • How do we recruit vendors to test this profile?
    • We have already reached out to CAP, so we have users aware of the profile but we potentially should reach out to the vendors = need CPOE vendors and LIS vendors
    • For the Connectathon we need to develop the test cases around the use case we want to have tested; we need user input on that. Riki/Raj will talk about it on the webinar on Aug 22 and will be promoted to the technical team at IHE
    • EPIC may handle both sides (order placer and order filler), so we need to reach out to them
  • New capability in the lab / both clinical and AP – maybe publish a paper to get more attention around this profile (possibly in Archives of Pathology and Laboratory Medicine). Jim can draft the article and Riki will help.

Digital Pathology - IHE and DICOM Digital Pathology Profiles Diagrams PowerPoint

  • White paper – brief look:
    • The Evidence Creator currently sends results to image manager, but the results of the evidence creation have to be conveyed back to the LIS.
    • Digital image K167 proliferation index is a percentage number (e.g.55%). This normally gets stored in the image management software = evidence creator or in the LIS
    • Annotation is on the image (that is how radiology does it now) inside the DICOM object as metadata – we need to support that; may not currently be the case
    • The result of the analysis is included into the report – so we need to have the evidence be treated like a result. And, where the specimen is the WSI – we need to support this, too
    • Update the global picture for the white paper – digital asset (which covers the image AND the additional metadata, including the evidence around the image)
    • Update query into the image display to read digital asset instead
    • Human annotation on the image is Evidence Creator – but image display is needed for this functionality (so that system may represent both)
    • DPEC has 2 actors: Evidence Creator and Image Manager / Archive
    • DPIA in image display role only needs to implement the image query and retrieve digital asset
    • Store the numeric result not as annotation on the image using these profile = DP analytical Data exchange
    • We can use LAW for that / Evidence Creator = Analyzer:
      • Create that as an alternate image showing the AWOS on digital asset from Analyzer Manager to Analyzer
      • AWOS result from Analyzer to Analyzer Manager, which can include a link to an image as well as all the results
    • LIS vendors currently only want to link to the image – no display of the image (is in WSI viewer software; WSI creates the evidence into the image) = how would the link to the image get back to the Order Filler?
    • Per Raj: While you look at the slide, you are dictating what you see. The human being plays the role of image display. The application recording what has been dictated plays the role of the Analyzer for sending the result upstream. You are creating a structure for the report in the dictation; there are key words in that, so every keyword becomes an OBX
  • IA:
    • What should happen when we re-scan slides? we get multiple IDs for the slides- need to ensure those ID schemata do not collide
  • Issues review:
    • Issue#10:
      • DICOM can only do 3 channels (Color calibration = ICC) – but in fluorescent you can have 7 different colors. Should this be out of scope? Per Raj, it is a big part, so it needs to be in scope (HER2 FISH, biomarkers etc.). Agreed to leave this issue open for now as it is not clear why the ICC would not be able to cover the different colors? (But calibration is not done for FISH today, because colors are so distinct)
    • Issue 11:
      • There are no standards for glass slide production to ensure reproducibility of algorithm review (could be on the glass slide processing OR in image acquisition) - this will be BIG problem – currently can only train the AI algorithms with multiple different images vs providing color calibration to standardize the image, so that the resulting images would be normalized and provide greater consistency
      • This will be an open issue until the industry comes up with a standard to normalize digital images. Once that has been described, then this profile can address how to convey the required data elements as part of the data exchange (or DICOM may address this). Need to be able to get consistent color in paint; maybe can use something from the AEL standard (However, this is not specific to this profile)
    • Issue 12: (moved to closed)
      • Need identifier type and assigning authority – can use either SPM-31 or OBX
        • DisplayID (Nick) to sort images in the viewer – either by algorithm from the image manager or manually by the user
        • Need the code, finished definition and description we want and then submit harmonization proposal (Riki)
          • IDSO = Image Display Sort Order Number
    • Issue 13:
      • Digital scan IDs must be unique, so we need to define the identification schema that uniquely identifies the new digital image = Digital Specimen ID (SPM-2.2 in LAB-82) – could be SPM-2.1 in a consultation workflow, where that created WSI is being sent for further work
    • Issue 14A:
      • Closed issue – review the diagram
    • Issue 14B: (moved to closed)
      • Study instance UID = 1:1 mapping to the case number = surgical procedure on the patient but only in relation to the lab the specimen is sent to = assigned by the filler = Order number
        • Usually the LIS creates the case number = Acquisition Manager
        • OMI message has IPC, but no SPM
          • Riki to check in with Co-Chairs of HL7 Imaging Integration (II)
        • Francois suggests using the ORC-38 to carry the filler order group number = Accession number = DICOM object ID = 0008,0500
        • Will use OBX-5 for now
    • Issue 15B:
      • Is this sent via the HL7 message or should it be handled ONLY in the DICOM object? This is to create fidelity of the image data; it will be in the DICOM object. But, do we need to send it separately in the HL7 message?
      • Should we include it as optional element and provide it in the Appendix mapping? If we do this, we would need to also include the description when to use it under what circumstances. We will have DICOM WG26 provide guidance
      • May be incorporated in a future option of the DP profiles
    • Issue 17:
      • QC identification of the glass slide: once an image is taken it will be the same specimen, so it is hard to identify the control part
        • In SPM-11 there is the Specimen Role, but that only works if the glass slide is the container with more than one specimen
        • We are including the parent specimen information that can have the SPM-11 properly filled out, BUT not sure how to identify the position of each parent specimen in the image. Then the specimen ID is the same for the two specimens on the glass slide in practice. BUT in each lab the control would ALWAYS be in the same place on the slide
      • We do NOT see the label image as often the scan will exclude the label of the slide and take a separate label image (using OBX-5) and OBX-5 for the QC section of the slide and the tissue area separately for standardized slides
      • Negative control is always a separate slide – so that can be taken care of with existing SPM-11
    • Issue 18:
      • Use OBX-18 for all OBX segments that are produced by the scanner
        • In LAW we use:
          • First repeat for model number
          • Second repeat for serial number
          • Third repeat for firmware
          • Manufacturer is the assigning authority on all of these
        • We are not really supposed to assign meaning to the order of elements in fields – and in the later versions of HL7 we have the PRT segment to describe the device with all these attributes – but we won’t use it here
    • Issue 19:
      • Provide the DICOM to HL7 DAM mapping as Appendix – we have the solution, but have not done it yet
    • Issue 20:
      • NTE for the OBX it belongs to, if there is such a relationship – unless it is a result, in which case it should be a separate OBX
      • OBX for vendor comment otherwise
      • REALLY, it should be better defined depending on the content
    • Spreadsheet updated - Raj will make these updates on Google
    • Next Steps = Renumber the issues in the closed section
  • Review of the profile publication timeline – Publication Timeline Available Here
    • SET profile question: Do we want to support ACKs – yes only commit level ACK, no application level ACK. Copy ACK Choreography from SSU^U03 in Chapter 13
    • PaLM TF CP:
      • LPOCT profile update: LAB-31 – sends the results from POC result generator (analyzer) to POC manager, related to QC results. We need to clarify the existing description of the conditional use for level_cd and cal-ver_repetition
      • In Japan they are using vendor specific protocols, so not tested in Connectathon = only tested LAB-32; same in Europe, as the vendor provides the complete solution up to last year, which is why we have this CP
      • Motion to accept the proposal as presented – Riki Merrick, Ralf Herzog, no further discussion, against:0, abstain: 0, in favor: unanimous
    • Updating the big picture of all profiles for the profiles that changed for comment to Trial Implementation = for LCC done – maybe update TMA to TI, if we know we will get there in time

SET Profile CR - Link: HL7 CR SET with Comments (Revised)

  • Specimen with package:
    • Do we need SHP?
      • If we do use this, we will have duplication with PRT / EVN attributes. We would not be consistent with other messages in SET
      • The only attribute we do not cover is number of packages
      • Shipper information can be carried in PRT as well
      • Shipment ID = if we need to have the transportation tracking number then we must use it
      • If we use SHP, then we MUST also have SHP-4 = Shipment Status Date time. Due to the context we would also need the SHP-3 = shipment status (which has these values from HL70905: INV Inventoried PRC Processing REJ Rejected TTL Triaged to Lab TRN In Transit). We would have to provide guidance how these status codes are related to the triggers we have for these messages
      • If we leave the SHP, we can only reject the shipment if we take it out package
    • How do we deal with Package of packages?
      • Make the Specimen group optional – but then we would have the situation that folks could send empty packages
      • SHP is maybe describing the package of packages. If so, then make the specimen group required within the package group
      • For the shipment acceptance we don’t need the detail inside it – so different message structure for shipment are shipment accepted or shipment rejected
      • We can then send accept or reject event messages for each opened package until we get to specimen, which we have in the other trigger events
    • Decision: we will remove this set of package / shipment centric triggers and delay that topic for a later CR, if people ask about it
  • Identification / Re-identification
    • Update the information on the re-identification Entity and De-identification entity – ALSO NEED CODES FOR PRT-4 HL70912
    • Where do you put the new specimen ID?
      • SPM-2
      • SPM-3 is the old specimenID; Create an example for this one
  • Specimen Disposition
    • Reason retrieve => ENV-4– part of the general attribute table
    • Reason archived => EVN-4 – part of the general attribute table
  • Specimen Processing should be before derivation
    • Can report on the processing first (on the parent specimen) and then need to statement that you now have a new specimen = child specimen
    • Can process a specimen without creating a new specimen
    • Use of OBR-4 to identify the procedure step
  • Post discussion thoughts: Will need to think about how to indicate successful processing vs unsuccessful, as OBR-25 is NOT the right place for that. Maybe it is best to create business triggers like we do for successful specimen collection/ unsuccessful collection

Wednesday, May 29, 2019

LCC Review

  • A few changes applied:
    • p. 6, Open Issues and Questions: A paragraph was added describing potential future work related to more flexible types of communication.
    • Line 191, sentence added to indicate that LAB-6 and LAB-7 are independent. Fix the diagram so not to have arrows on the connector lines (Riki will do)
    • Line 203, edited to indicate separate sets of use cases for LAB-6 and LAB-7.
    • Line 225, edited to add “independent.”
    • Lines 316-320, minor edits to indicate the independence of LAB-6 and LAB-7.
    • Line 386, 390-394, text added to describe the use case related to insurance pre-authorization. This fits into #7 a bit better than #6, which would have been the other possibility.
  • Next step for Riki to send to Mary Jungers once the diagram is updated. The goal is to publish the first week in June

TMA - Blood Transfusion Administration Workflow PowerPoint and IHE PaLM Technical Framework Supplement

  • HL70513:
    • TS = Transfusion Started
    • TI = Transfusion Interrupted
    • TX = Transfused = normal end to the transfusion
    • Riki to prepare harmonization proposal for HL70513 (include definitions for all codes)
  • BTX-19 - make RE (we do not want to make it conditional on BTX-11). HL70515 is only concept domain – no suggested values – how do we come up with codes for this?
    • Need to check in FHIR
  • BTX-14 = performer – should be the nurse = make RE
  • BTX-15 = verifier – should be the person that verified that the match was = make RE
  • In Belgium and Netherlands both are needed. In Japan, we also need 2 persons but there is no rule who that must be (could be nurse/nurse; nurse/doc; doc/doc). This information is recorded on a paper report that is recorded, also in the EHR
  • BTX-18 – make RE – HL70514 is user defined – HL7 base standard has some suggested values – we could start with those
  • Blood product label for Japan is locally defined
    • Blood type
    • Blood product code
    • Lot number = assigned per donor = how is it made unique?
    • Date of collection
    • Date of expiration
    • All the above are in the 2D barcode
  • In Japan, no pooled blood products are allowed for plasma, but for blood product components like Albumin
  • There are four vendors working on this profile: MIPS, University of Loewen, EPIC and a smaller company
  • In Japan, is the blood transfusion documenter (Bedside) separate from blood product filler = Blood bank LIS?
    • In Japan the EHR vendor captures the transfusion documenter – the EHR vendor provides the interface specifications to use – can those be shared? They can share the kind of elements being shared, but not the specification, as it is proprietary = already in production
  • Next Step for TMA:
    • Francois will send the latest version to Filip and Riki
    • Riki to create harmonization proposal for HL70513
    • Filip will draft and share with Riki to review and update. We plan to review final version on July 10 call
    • Japanese blood bank system vendors will send the data elements currently exchanged in production to Filip

Wrap up Link: Publication Timeline Available Here

  • Francois’ summary document:
    • LCC: proposed for Trial Implementation for June 3 (no change)
    • TMA: proposed for Trial Implementation publication. We are aiming for July 16, review on July 10 call
    • PaLM TF V10 publication aim for July 1 (no change)
    • DP assets:
      • White paper is almost ready – aim for publication for June 15
      • IA – aim for publication for Public Comment for June 24; 1-month review period (closes July 29)– aim for Trial Implementation for Aug 29
      • OR = Ordering and Reporting for ALL LABS – updates for LAB-1/PAT-1 and LAB-3/PAT-3 from LTW and APW; should look into requirements from ILW – focus on any lab ordering, not just AP – but also clinical and micro support, not just inside hospital focus on use of ORC-4 and ORC-38 respectively to provide linking across multiple workflows – aim for publication for Public Comment for July 16 – aim for Trial Implementation for Aug 29
    • SET:
      • Need to still drop out the specimen movement by package => resulting in renumbering of the trigger events
      • Specimen Re-identified
        • Where to put the new and the old specimen ID
        • New specimen = current ID = SPM-2
        • For previous Specimen ID send in SPM-30 with NEW identifier type code of Previous specimen ID = need to include in HL70203 harmonization proposal
      • Specimen processing
        • Will use OBR-4 to identify the processing step
        • Cannot use OBR-25 for status, so make this NEW triggering event: Successful processing event/Unsuccessful processing event
      • Next Steps for SET:
        • Alessandro will update draft per discussion from F2F and share with Riki by end of this week
        • CR review in HL7 OO on June 13 call (Riki will re-introduce the topic on June 6)
        • Complete Volume 2 for review of the entire supplement by IHE PaLM for Special call on July 24 9 – 10 AM EDT
        • Publish for Public Comment for August 7 (need before Sept 16). Process the comments for F2F in November and then publication for Trial Implementation right after, i.e. Nov 22
        • Had some folks from Belarus interested in implementation for next Connectathon 2020
  • Next F2F meeting in Chicago
    • Nov 13 -15 at CAP headquarters in Northfield, IL
    • Next F2F after that may be in Paris or in Belgium (Gent). Filip will look into the possibility of hosting it in May (NOTE: HL7 is May 15 – 22 in San Antonio, TX)
  • Webinar planning
    • August 22 12 – 1 PM EDT. Riki and Raj will give the webinar.
    • The webinar will reflect the above workplan and to solicit interest in the NA Connectathon; the main topic of the webinar will be Digital Pathology
  • Moving to FHIR:
    • Order catalog
      • Building new resources for services – have been testing at Connectathon for lab (PHAST, Quest and Russian company)- -but resources are still at maturity level 0 – plan to upgrade to 1 for this year
      • HL7 catalog calls are weekly on Fridays 12 – 1 PM ET
    • LIVD mapping
    • The rest of lab workflow is not yet well worked out
      • Observation resource is normative
        • But has multiple ways of handling parent child linking – plan to work on this this cycle
      • The guidelines on using DiagnosticReport and Observation together still need clarification. It is not completely clear when an Order Filler should provide its results using only Observation resources and when it should wrap these Observations within one (or more?) DiagnosticReport resource.


Meeting Adjourned at 12:05 PM local time.


Thank you to JAHIS for hosting the meeting!