Health Ontology Mapper
Space shortcuts
Space Tools
Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 48 Next »

OntoMapper Minutes

3/15/2011

Participants: Rob Wynden, Vijay Rayanker,

1)

2)


2/23/2011
Participants:  Prakash, Nive, Harsh, Vijay, Aaron Mandel, Peter, Mark

  1) UAB working on bringing in Lab data into i2b2

  2) Vijay working on making UETL accessible from command line

  3) Prakash working on port of View creator onto Oracle


2/15/2011

Participants: Prakash, Vijay, Ishani (UCLA), Peter, Curtis (UAB), Nive (UAB)
 
* UCLA has minor issues with UETL imports probably due to wrong usage, another screen-share to be scheduled.
* Vijay is working on UETL updates.
* Diagnoses and procedures have been implemented at UAB.
* View Creator deployed at UCSF with an additional view for Visits.
* Discussed on bringing in Lab data into i2b2 at UAB.

2/8/2011
On the call: Rob Wynden, Ketty, Prakash, Vijay, Nive, Harsh (UAB), Ishani and Khy (UCLA), Peter B, Curtis H

1) We had a demo for U Chicago and sent them the executable

2) UAB i2b2 in Production had trouble but now it is working correctly now

3) UAB still using demo data ... need to do the Oracle Server SQL load bulk load step

4) We need password for SVN from Aaron Mandel.

5) UCLA needs the biostats view

6) Vijay working on making command line execution

7) Prakash is working on the View implemention/optimization

8) Ketty is working on Protege mapping process into BioPortal

9) We had a very successful with Shawn's team


2/1/2011

On the call: Rob, Prakash, Ketty, Vijay, Nive, Harsh, Dave Hau (from NCI), Khy, Shani (UCLA), Adrian (U Chicago), Mark Weiner, Curtis

1) UCSF created a View structure.  This allows access to i2b2 from SAS, STATA, R and SPSS

2) Site updates

3) Ketty working with Lori at partners on the scripts.  Automation of the UETL concept paths.

4) There is a webinar on RxNorm and meds tomorrow

5) UAB having some build problems in production.  They are working through the AUG now.

6) Demo with Curtis


1/25/2011
On the call: Rob Wynden, Prakash Lakshminarayanan, Ketty Mobed, Aaron Mandel, Vijay Rayanker, Nive and Harsh, Peter B., Khy, Adrian from U. Chicago, Mark Weiner, Curtis H.

1) Khy at UCLA has i2b2 running now and now they need ETL help.  Now installing HOM and working Vijay.

2) UAB is working on main production database now.  They are configuring the VM in a more efficient manner now.  Using test data still for now.  After test will use real patient data.

3) Vijay has made the changes to support not generating proxy MRNs.  Change made by UAB and posted in main code base now.

4) U. Chicago has i2b2 running and needs UETL help with cancer registry.

5) Rob working with Matt Wyatt at UAB on submitting an admin supplement related to this work from UAB.

6) Still changes on BioPortal changes. Due this week

7) We have been running many maps off of BioPortal, Disposition and working on Formulary.

8) UAB is working on lab mapping now

9) UCSF working a stored proc that allows you to extract all data associated with a patient

10) We have a demo of HOM / BioPortal for Shawn Murphy's team today at noon.


1/18/2011
On the call: Rob Wynden, Prakash, Vijay, Mark W, Nive and Harsh, Ketty, Khy (UCLA), Aaron Mandel, Mark W., Curtis H.,

1) We are still waiting on BioPortal to make our bug fixes

2) Vijay has loaded all UCare data.  Allowing users to avoid making proxies of MRNs or Birth Date etc... for UAB

3) Nive: UAB is now working on loading real data (they had been using test data).  They had been working on a VM instance with Oracle ... but now they want to use the main database.

4) Vijay to start working with Khy at UCLA

5) Vijay will work on integrating UAB support for Views into UETL


11/16/2011
Participants:  Prakash, Peter, Vijay, Curtis
 
* Prakash is finishing up the map for Patient and Encounter mapping.  Another map that exclusively supports Patient mapping alone needs to be developed since UCare Discharge table has only patient reference.
* Vijay is loading UCare and AXIUM data.
* Peter talked about long and unpredicatbale execution time of i2b2 queries at his site.
* UAB is doing prep work on i2b2 install.


1/11/2011
On the call: Rob Wynden, Ketty Mobed, Prakash Lakshminarayanan, Vijay Rayanker, Aaron Mandel, Mark Weiner, Nive and Harsh at UAB

1) Further UETL testing at UAB.  They need to use proxy MRNs and proxy Encounter Numbers.  They did that and UETL is now working at UAB.

2) UAB has ICD9 and Labs In system now ... not mapped to main ontology

3) Prakash is working on Sybase IQ support last week.  We are blocked on NCBO BioPortal for their bug fixes.  We are importing all of our ontology there now...

4) Prakash will focus on LabNorm for now.

5) Vijay is loading more UCare data.

6) Vijay is making some enhancements after talking to UAB.  To make the proxy list show up in the GUI to make it easier to maintain.

7) Ketty asked Mark about ICD9 and ICD10 support.  Should we use an existing crosswalk or create our own?

8) RxNorm updates are not getting handled fast enough in BioPortal


1/4/2011
On the call: Rob Wynden, Aaron Mandel, Ketty Mobed, Harsh Taneja, Niveditha Thota, Vijay Rayanker, Prakash Lakshminarayanan, Mark Weiner, Peter Beninato

1) New site for our dev work is at community.i2b2.org

2) Harsh question on SVN site.  UETL code is read only.  Could not get HOM code however due to bad permissions on the SVN.  We will fix that for her.

3) At UCSF we are working on production issues (we upgraded Sybase to 12.7).  That is postponed for now.

4) BioPortal integration is now waiting on fixes from Stanford.  So we are stuck for now.  Ketty will make sure that we have a ticket opened at Stanford. The source terminology cannot be properly annotated at leaf nodes due to a bug in the BioPortal UI.  We also need private NCBO sites working very soon (for license restriction).

5) We have a joint lab meeting (UCSF and Stanford) at Stanford on Thursday of this week.

6) At UAB the usage of UETL is working well.  They seem to be clear of all exceptions and they can search for stuff.

7) Asma ICD9 455 and search by name in ICD9 tree (using Prakash's) and it works now!  So UETL is running at UAB.  BUT SEARCH BY CODE IS STILL GIVING AN ERROR.

8) UETL needs to have a check added for duplicate entries in the i2b2 table (this is a bug in UETL) - Vijay will work with them on this.

9) Ketty is working now with Peter's ICD9 mappings

10) Rob and Ketty are working on a paper about our integration with NCBO BioPortal

11) Regarding our work on ontologies... The NCBO Site: http://bioportal.bioontology.org/ontologies?filter=CTSA-HOM shows everything that our team has been working on so far.  HOM has a LOT of terminologies available here already.

12) Next week we will see a demo of the UAB ACE (Aggregate Cohort Estimator) query interface


12/21/2011
 Ketty, Prakash, Harsh, Peter, Curtis, Nevi, Aaron
 
Prakash – updating Sybase
 
Ketty – more uploads and annotations in NCBO BioPortal. One UCare table (DISCH) is there. While proprietary, it was put there to the public site for our group to review and give comments back to Ketty. We are still waiting on the public site from Stanford.
 
Nevi (UAB) – have used UETL to import files to the i2b2 sql loader. They encountered some issues with some nodes and database. Also an issue with the drag and drop option showed nulls at the tool tip. They were able to resolve most issues. Prakash discussed some of the items with them, but some of the unresolved issues will wait until Vijay comes back in 2 weeks.
 
Curtis (UAB) – while they have to wait on the UETL issues, they want to get started on HOM. Prakash will set up a call for tomorrow (12/22) pm and get things started for them.
 
Aaron – HOM wiki has been relocated to Harvard (wiki.i2b2.org). Only he, Prakash and Rob have access to it right now. Won't be available until Rob gives his okay.


12/14/2010
On the call: Rob Wynden, Harsh, Ketty, Peter, Prakash, Mark Weiner, Curtis and John from UAB also

1) Ketty: we have BioPortal working but just hit a snag ... NCBO is working on fixing it.  UETL of source data for subsequent mapping.  This requires annotation on NCBO for mapping after the terminology is pulled down by Shawn's new BioPortal terminology pull feature.  The target ontology also needs annotation.  But the source terminology cannot be properly annotated at leaf nodes due to a bug in the BioPortal UI.

2) Harsh is wondering about the patient and visit data ... how do they map the rest of the info.  Diagnosis, procedure and lab results?  Now what is the next step.  They need to run the bulk import files.

3) UAB seems to have some trouble with re-use of proxy MRNs.  We will forward a question on this to Vijay...

4) Prakash, the POC for BioPortal integration works.  We are focused on uploading more terminologies now.

5) Diagnosis map for UCSF being created for UCSF too

6) Peter: has an Oracle dump of ICD9 codes from UMLS (latest ones).  Ketty really could use the list of invalid mappings so that we can work that into the BioPortal maps for ICD9.  Peter will supply them as an Excel or CSV file.  Very helpful.


12/7/2010

On the call: Rob Wynden, Ketty Mobed, Mark Weiner, Aaron Mandel, Peter B, Curtis, Harsh & John at UAB

1) Vijay is out on paternity leave until January.

2) We have Bob Fink working on updates to UETL now... he is automating the repeated execution of the tool and making it keep track of the generation of bulk files, their xfer and loading

3) We have maps running now on BioPortal and OpenMDR.  It works

4) We have a joint lab meeting with the BioPortal team in January

5) We have a meeting with Shallaby's company later this week

6) Shawn's team is working on the terminology load into i2b2

7) Harsh is still working on patient and visit population

8) An update to the healthontologymapper.org site is waiting for our approval

9) Ketty will update our contact to the public site

10) Rob to contact Nigam for the 15th deadline of a podium


11/23/2010

On the call: Rob Wynden, Aaron Abend, Prakash, Harsh Taneja, Mark Weiner, Vijay Rayanker

1) Met Jim S and Nigam about NCBO integration

2) Vijay is loading more Axium data via UETL as well as GE UCare

3) The CRC algorithm is going to leverage a trigger on the staging table where we UETL data out of

4) Vijay and Harsh have been working together.  The source data at UPenn is in Views.  UETL will be updated to support source data in Views.

5) Prakash is done with patient and encounter HOM mapping.  Links tables that don't have encounter and patient data... the map will look up the right values and update the patient and encounter table.

6) Prakash working on NCBO integration.  He has everything to execute the maps from the bioportal.

7) Ketty has updated a lot HOM terminologies to BioPortal.  The discharge is now all up on bioPortal and the others are still waiting for finished curation in Web Protege.  The NCBO will house all of our ontologies in bioportal as "CTSA Data Sharing".

8) Harsh will make the change to support Views - and work in with Vijay


11/16/2010

On the call: Prakash

ROB NEEDS TO ENTER PRAKASH's MINUTES HERE...


11/9/2010

On the call: Rob Wynden, Vijay, Prakash, Mark Weiner Ketty Mobed, Harsh Taneja (UAB), Peter, Curtis, John Osborne.

1) Rob arranged a meeting with Jim S.

2) Natasha Noy announced an interim deployment of mapping master.  Ketty is testing it with the discharge codes.

3) Our ICD9 views (alternates to Harvard view) called ICD9 Code view created by Ketty

4)  We are working on CPT codes now.

5) The NCBO think the REST interface for OpenMDR will be fixed by Thanksgiving

6) Loading the encounter and patient mapping - post UETL.  We are creating a workflow for that

7) Loaded a LOT of data using UETL into i2b2.  Also starting in on CRC algorithm.

8) Hersh wants to try out UETL and HOM.  Will interface with Prakash on that.

9) Curtis talked about a new sub-project we are initiating that will allow cQL to tunnel URI's to make it compatible with SHRINE and HOM.

10) Curtis and John are exploring the use of the SHRINE connecting cells on each end of a grid.  We would connect the ports of SHRINE cells as a data service to cQL using 2.0+ of cQL.  We would use authN/authZ project to handle security on TRIAD.  Our end goal is to simplify deployment by leveraging caGRID security for large SHRINE networks.  This would also enable non i2b2 connections to i2b2 based grids as well.

11) Mark will help Rob out on the RxNorm paper.


11/2/2010

On the call: Rob Wynden, Curtis and John (UAB), Vijay, Prakash, Aaron A, Aaron M, Peter B

1) Rob arranging meeting with Jim S.

2) We have the HL7 Dispo and UCSF TSI checked into NCBO BioPortal

3) We WERE able to create a mapping and get it working!  We are dealing with some minor bugs in BioPortal.  Stanford is working on fixes within the next week or two it should be fixed.

4) Can OpenMDR pull it's codes from BioPortal?  Or can codes be pulled into OpenMDR from a flat file?  It's cumbersome to enter codes otherwise.

5) UETL just using the tool to extract lots of data for testing.

6) Prakash will be working on infrastructure to run HOM interpreter right after UETL load.

7) We talked about small cell results (HIPAA) masking


10/26/2010

On the call: Rob Wynden, Aaron Mandel, Aaron Abend, Prakash, Curtis & John (UAB), Vijay, Deb and Brian

1) Rob to follow up with Jim Shalaby about commercial map support

2) Rob to set up a "Related Project" site on the i2b2 forum.

3) Recom has a new public site for HOM

4) Several small maps in Web Protege curated by Ketty

5) Deb and Brian sent us a 1.3 build for the query tool

6) Deb and Brian are working on the 1.5/1.6 i2b2 bug... it returns the wrong results (so we are supporting 1.3)

7) UCSF is creating regression test scripts for i2b2 (using ALODS/OSIM)

8) The posted HL7/Dispo is good enough for us to finish the NCBO BioPortal integration

9) The 1.3 build exe will be posted to the HOM wiki (after UCSF has posted it to production) - Deb and Brian to be referenced as the authors from now on...

10) UETL is still running the GE UCare EMR data.  We should be done with the main tables by end of day Friday

11) The CRC work is done we can then focus on the CRC work (support for updates)

12) The LabNorm work will restart in a week or so

13) Multiple tomcat instances for SHRINE should work fine... we talked about this as a way of creating restricted data views over SHRINE (they all refer to the same JBoss instance)


10/19/2010

Participants:  Prakash, Aaron Mandel, Vijay, Ketty, Peter

1) Ketty is working on getting Harvard Demographics ontology onto Web-protege.  Discharge Disposition which is a proprietary ontology would go into the normal version of the protege.
2) Vijay is working on UCare extracts ... this is a huge test of the UETL process (every byte of a large EMR)
3) Vijay and Prakash would be meeting with COHRI staff to build Dentistry (Axium) ontology and also to create mapping between raw Axium data and the Dentistry ontology.


10/12/2010
Participants: Prakash, Aaron Mandel, Peter, Vijay, Ketty

1) Prakash talked on last week's Stanford meet and UCRex demo
2) UETL - Vijay is extracting Axium & UCare data
3) Webprotege - Aaron to send link on Harvard Ontology to Ketty 


10/5/2010

On the call: Rob Wynden, Aaron Abend, Peter B, Vijay, Prakash, Ketty

1) Stanford meeting on Thursday: NCBO BioPortal team meeting about integration with i2b2 and HOM

2) Prakash is working on the support for i2b2 1.5 (also it's on Oracle and not Sybase).  Should we be using 1.6 now?! (It has Encounter implemented)

3) Lots of researchers are signing up for service at UCSF now.  We talked about the training that Ketty is doing.

4) NCBO BioPortal: waiting for Web Protege account so that we can enter the Discharge map.  Also working on mapping out UCare tables for that terminology and ICD-9.  BioPortal has converted our ICD9 to OWL with Martin's Web Protege Mapping Master - now we can check that into Web Protege.

5) We need a commercial vendor for maps

6) Vijay is running LARGE UETL loads.  Current extract is about 15 million records with a total of 65 million.  Full UCare is about 250 million records.

7) CRC is still just getting started.


9/28/2010

On the call: Rob Wynden, Ketty, Prakash, Vijay, Aaron Abend, Peter,

1) Vijay is just in process on the CRC implementation

2) Using UETL for Axium for Dentistry. Loaded about 10million records into i2b2.  Demo for this call next Tuesday at 11.

3) Prakash, is moving us to i2b2 1.5 using Oracle.

4) NCBO integration with HOM: Ketty is working with the Web Protege team to get our terminologies checked in.  This will allow HOM to run from BioPortal.

5) We will shoot to get Disposition actually running from NCBO BioPortal this week.


9/21/2010

On the call: Rob Wynden, Aaron M, Aaron A, Tim, Peter, Ketty,

1) CRC algorithm is in progress, we will likely need advise on this next week

2) Ketty is working checking in some terminologies into NCBO Bioportal (converting ICD9 Disease Codes to OWL).  Disposition is just starting and Prakash would like to use the Disposition map to prototype the BioPortal-OpenMDR integration.

3) CPT is also a new map that we need to complete

4) ICD9 Procedure Codes are not covered yet.  Asked Phillip Reeder for their ICD9.  We need a ICD9 to ICD10 map

5) We need visit for HOMERUN ... UETL has that done now

6) Ketty has protege 3 installed (needs plugin supplied by Stanford)

7) Prakash is still working on LabNorm


9/14/2010

On the call: Rob Wynden, Peter, Vijay, Prakash, Phil, Becka

1) We did a live demo of HOM (UETL and Interpreter) at the CICTR conference and it went well

2) NCBO BioPortal work is waiting on the creation of our ontology on the BioPortal in WebProtege.  Rob will talk to Ketty and get started with Disposition

3) TRIAD OpenMDR: HOM is calling into the API on the OpenMDR dev site.  We are able to get Disposition codes from that site.  So it works and we are ready for starting with BioPortal

4) Vijay, is working on a very large ETL load for Axium on the CHORI grant.  We sent some files to UT but we don't have confirmation.  Also we extracted an additional 5 tables (we ran out of memory)

5) Phil and Peter are checking about local interest in the HOMERUN grid


9/7/2010

On the call: Prakash, Peter, Tim, Becca, Vijay, Mark

1) UETL status - Vijay is extracting the Dentistry data of Axium

2) OpenMDR status - Prakash was able to define Discharge Disposition codes on OpenMDR Dev site

3) HOM google groups issue with regard to file downloads and missing mails to be investigated


8/31/2010

On the call: Rob Wynden, Prakash Lakshminarayanan, Vijay, Mark, Deb, Brian, Nick, Phil

1) The name TCH Denver Plugin is the new name for the query by value code

2) UETL code is check into subversion and executed posted to public site.  Docs are posted too.

3) REST APIs work with BioPortal and OpenMDR.  Prakash has now been trained on this.  We will use Disposition to prototype the integration.

4) Vijay has trained Hari on UETL.  UW would install it in 2 or 3 weeks.

5) TCH Denver is having trouble with i2b2 1.5 table viewer (not related to their plugin)

6) Phil will check in updated drawing for the CRC algorithm


8/24/2010

On the call: Rob, Nick, Prakash, Aaron Mandel, Brian, Phil, Vijay, Tim

1) Disposition map works at UCSF.  Now we need to test at the other CICTR sites (they are live with a manual mapping now...)

2) Vijay's code not checked in yet. He is doing that now.  It's working fine and will check in now.  Kaiser next up on UETL install I think.

3) REST API's for BioPortal are working now.  For OpenMDR we have not yet figured out how to call it.  Prakash needs help setting up Point-to-Point maps.  Rob will arrange call with Rakesh Dhaval.

4) We walked through Phil's last edits to our CRC algorithm for next version of UETL


8/17/2010

On the call: Rob Wynden, Prakash, Vijay, Aaron Mandel, Tim, Becky, Nick, Brian, Peter, Aaron Abend, Mark, Marco

1) Code complete (encounter, demographics) except does not have i2b2 metadata table and testing.

2) Running Disposition maps for CICTR/HOMERUN on the data loaded with UETL.  It worked and runs in only a few minutes.

3) Some of the source disposition codes were not defined in Davera's mapping.  Did we not include them as possible values?  We should have used the UDenver plug-in (after UETL) to find list of all possible values.

4) There are a ton of new little projects starting up around the NCBO BioPortal integration

5) The UDenver Plug-in is REALLY useful.  We recommend it.

6) UPenn sent over their Provider mapping code.  UCSF will derive a new version based on that.

7) We will return to the CRC generation architecture next week.


8/10/2010

On the call: Rob Wynden, Vijay, Nick, Prakash, Becka B, Marco, Aaron M, Gaurav, Peter, Mark W., Aaron Abend

1) Vijay is code complete for Visit and Encounter (with Proxy MRN) - testing this week

2) NCBO Bioportal Training session

3) Prakash dev efforts - web services call from HOM interpreter to Bioportal - point to point mappings are first on BioPortal - we will move onto other map types as well

4) Mark will have Alice send Prakash the PHP code for Provider


8/3/2010

On the call: Rob Wynden, Prakash Lakshminarayanan, Tim, Nick, Marco, Aaron Mandel, Gaurav, Peter, Mark

1) UW test of new HOM (Prakash and Gaurav)... still getting access privs for Gaurav.  Should these results be included in the RxNorm journal paper?

2) Prakash finished his code cleanup (re-factoring).  New version is already posted on healthontologymapper.org...

3) Kaiser coming up on the RxNorm script this week (CPT mapping is also of interest to them)

4) Nick wants to create a HOM VM with everything prepackaged.  Gaurav will work on it.

5) NCBO BioPortal wants to house the HOM scripts.  NCBO wants to help with the NLP work.  A mapping workbench is a possible collaboration too.

6) Disposition mapping issues.  We will sync up both HOM and non-HOM sites for this test as part of the final CICTR Aims with UW and Nick.

7) Start in with Provider map with UPenn (Mark Weiner) to get that work into a formal HOM map.  How to we query on a Provider at a particular Clinic?

8) Clinical Informatics Symposium paper deadline is soon (** should this be a paper on the NCBO integration work? Rob to talk to Nigam **)


7/27/2010

On the call: Rob Wynden, Nick Anderson, Prakash, Marco, Tim, Beck B, Mark, Gaurav Hardiya (UW/San Jose State), Vijay

1) Renewals

2) UW Implementation of enhanced HOM with rerun - working with Prakash to get it running this week  on the new version - also creating Centos VM distribution

3) Prakash code re-factoring almost done.  Code cleanup - nothing major but necessary - then Disposition maps

4) Provider ontology work - Nick requests that we post the field over-rides to the AUG list

5) Vijay's work - Patient dim and demographics are all automatically generated now and sync'ed with Obs Fact.  Now working on Encounter for Visit Dimension


7/20/2010

On the call: Rob Wynden, Prakash Lakshminarayanan, Tim Thimman, Becka Barren, Rebecca Hazen, Nick Anderson, Aaron Abend, Aaron Mandel, Mark Weiner, Peter B

1) Update on Provider mappings with UPenn, assign doctors into practices and practices into departments to set this up - UPenn's starter set for the Provider Hierarchy is posted on Harvard site

2) Davera has just recently finished the Disposition mapping

3) HOM update for RxNorm - to eliminate any possibility of duplicates that is scaleable - so we are using the Provider ID column to concatenate what's necessary for key - testing at UW is today

4) UETL - same status - shooting to have Harvard ontology automatically populated by UETL by the first

5) NLP project is launching - Michael Bobak will join us on weekly calls perhaps starting next week


7/13/2010

On the call: Rob Wynden, Vijay Rayanker, Prakash Lakshminarayanan, Nick Anderson, Peter B., Deb Batson, Brian, Rebecca Barron, Phil Ng, Rebecca Hazen, Aaron Mandel

1) HOM Interpreter status and deployment at UW - Rochester is working now and we have yet another performance test - will install that at UW using 1.4 VM from i2b2.org

2) UETL Status - next rev (with Patient and Visit) feature complete at end-of-month

3) Support for updates in UETL - will start implementation at end-of-month

3) Talk about new mappings for Disposition, Chief Complaint (reason for visit) and Provider.  UDenver will look at Harvards papers on using Chief Complaint data

4) Talk about NLP filter for UETL to handle Clinical Notes (Pathology notes, operative and procedure notes, radiology reports, clinical notes with focus on admission history and physical notes and discharge summary). Daily progress notes may be too hard for now...


7/6/2010

On the call: Rob Wynden, Marco Casale, Tim Thimman, Vijay Rayanker, Rebecca Hazen, Prakash Lakshminarayanan, Aaron Abend, Mark Weiner

1) Performance issue with Rochester.  The cache is using too much RAM.  They are increasing the heap size and trying again.  We are eliminating duplicates on write at Rochester which is a different approach than UPenn (Rochester has a primary key on Obs Fact whereas UPenn did not).

2) Most likely we will use a compressed - in-memory DB to filter duplicates before we write to the bulk import file.  We are look at the TRIE algorithm right now.

3) UETL, Vijay is working on changes to allow population of the demographics tables in i2b2, and using GUIDs.


6/29/2010

On the call: Rob Wynden, Aaron Mandel, Marco, Tim, Rebecca Hazen, Becka Barren, Prakash, Vijay, Peter, Nick Anderson, Hari Rekapalli, Mike K., Aaron Abend, Mark Weiner

1) Rochester got the RxNorm script running already.  We are making our final run now with all optimizations turned on ... we are hoping for a 4 hour completion time ... then we can finally publish the RxNorm paper

2) UETL Design and Implementation Status - Our design for support for updates (thank you for such a large contribution Phil Ng) is ready.  Vijay will begin work on this soon ... see below

3) Mapping workbench code-base decision - we will return to the java code based on UCSF's design written by John Deshazo at UW, Mike K. has the code in the Recom SVN repository

4) Vijay's status on UETL - adding functions for fetching all demographics, Patient & Visit dimensions.  Then he will begin on implementing support for updates (via ProxyMRN and CRC)


6/22/2010

On the call: Prakash, Nick, Vijay, Tim, Aaron Mandel, Peter, Mark

1) Prakash updated the HOM redesign and interim build for Rochester

2) Vijay is working on ProxyMRN table pertaining to UETL

3) Mark brought up RxNorm hierarchy issue wherein he wanted an ontology / taxonomy other than current ordering of drugs.   The group agreed to the need for new hierarchy but no solutions were proposed.

4) Tim would perform HOM install at Rochester before this weekend.


6/15/2010

On the call: Rob Wynden, Phil Ng, Prakash Lakshminarayanan, Matvey Palchuk, Brian Eliason, Deb Batson, Marco Casale, Tim Thimman, Peter Beninato, Mark Weiner, Alice Livshits

1) High level diagram is complete, focus is now on Phil's detail diagram

2) We reviewed Phil's low level diagram

3) Okay, we are now ready to have Vijay go ahead and begin coding this feature

4) UPenn has installed the latest HOM and it seems to run well (they Did use the bulk import method and it seemed to complete in only 4 hours!)

5) HOM will be moving to i2b2 1.5 now - we talked about all outstanding issues and what it will take for 1.5 support


6/8/2010

On the call: Rob Wynden, Prakash Lakshminarayanan, Vijay Rayanker, Nick Anderson, Aaron Mandel, Phil Ng, Tim Thimman, Marco Casale, Peter Beninato, Justin Prosser, Hari Rekapalli, Mike K. Mark Weiner

1) The lab norm design doc is done and a deployment plan has been prepared by Rob.  This will make planning easier for Prakash.

2) UDenver plug-in has been released into production at UCSF to help field customer data requests

3) We continued with our group discussion on the CRC algorithm to allow i2b2 to accept updates.

4) We may be able to identify old records using the Archive Obs Fact table.  We could modify the UDenver plug-in to use that properly.

5) We will walk through how we can handle deletions from the source record next week (Phil Ng will prepare a flowchart for us...)


6/1/2010

On the call: Rob Wynden, Marco, Mark, Deb, Brian, Tim, Phil, Aaron M, Vijay Hari

1) CRC checksum algorithm

2) Adapt UDenver plugin to i2b2 workbench to use Data Source or Update Date in Obs Fact

3) Recom to research how they are already handling update and deletions in the field (if at all).

4) We will continue on this path next week...


5/25/2010

On the call: Rob Wynden, Hari, Tim, Phil, Mark, Vijay, Prakash, Chris, Peter

1) Reviewed the CRC algorithm

2) Status Prakash on HOM release - performance improvements (uses disk writes if configured on a properties file)

3) Processing of NDC codes has been added RxNorm

4) Vijay status on UETL update - adjustments following experience on UW


5/18/2010

On the call: Rob Wynden, Mark Weiner, Prakash Lakshminarayanan, Deb, Brian, Tim T, Marco, Vijay, Phil

1) We talked about the limits of RxNorm (it's very precise and not good for general purpose queries)

2) We talked about Prakash's idea for integration between HOM for v3 - add a column to the fact table (and remove the concept path as requested by Mike K.)

3) We are doing a post mortem for UETL after the UW grant deadline was met today

4) We talked about bulk file generation in HOM - that's working now (will be point release) - NDC code support is in that too


5/11/2010

On the call: Rob Wynden, Mark W, Alice, Aaron M, Mike K., Deb, Brian, Tim T, Prakash

1) UCSF has completed our performance optimizations for the HOM interpreter

2) UPenn has been working on their HOM install to run on the formulary (RxNorm)

3) NDC's are specific to manufacturer.  This poses a problem for UPenn but a recent extension to RxNorm should address that

4) Rochester working on the i2b2 v1.4 install and trying to fix a bug related to that

5) We talked about the architecture of the new SHRIMP that Hari is working on

6) UCSF and Denver will collaborate to allow the Denver plug-in to extract/report on HOM Mapped data.

7) Rob to have the CRC Checksum Update write-up ready for the next call...


5/4/2010

On the call: Rob Wynden, Peter, Prakash, Mark Weiner, Alice, Marco, Tim, Phil, Deb, Brian, Mike K., Aaron M

1) HOM optimization to make it run fast on VERY large data sets

2) Implemented memory caching that improved performance by 300 percent but it's still not fast enough

3) We now are using a bulk export/import method to make it run even faster

4) UW is running HOM (RxNorm) on over 40 million records!

5) UPenn will be running on over 100 million records!!  They WILL need this optimization.


4/27/2010

On the call: Rob Wynden: Rob Wynden, Aaron Abend, Marco, Tim, Prakash, Phil, Maggie, Deb, Brian, Peter, Aaron Mandel, Mark, Alice, Hari

1) Vijay is loading formulary data with UETL, then moving onto reuse of MRNs

2) Prakash got caching in mapping interpreter, much faster now (UW and UCD running it now)

3) UPenn is working on the production install and run on formulary data.

4) UDenver may do a HOM install ... support

5) Hari studying SHRIMP code.

6) Talked about handling of updates - it was a big topic

7) Rob will write up the CRC algorithm for the team and distribute


4/20/2010

On the call: Rob Wynden, Marco Casale, Peter Beninato, Prakash Lakshminarayanan, Vijay Rayanker

1) UETL - HOM interpreter INTEGRATION should be done today some time

2) HOM Demo is Wednesday - over 100 people have signed up already!


4/13/2010

On the call: Rob Wynden, Vijay Rayanker, Prakash Lakshminarayanan, Aaron Mandel, Marco Casale, Tim Thimman, Deb Batson, Brian Eliason, Peter Beninato, Vincent Yip, Aaron Abend,  Yaorong Ge, Michael Horvath, Matvey Palchuk, Phil Ng, Maggie Massary, Mark Weiner

1) We talked about troubles getting the PM cell to work with the new i2b2 1.4.  The install is easy but it doesn't seem to work.

2) Vijay gave a demo of the UETL feature of HOM v2

3) UPenn is still creating the production HOM server (Alice is working with Prakash)

4) Status on SHRIMP upgrade with Hari

5) Prakash is helping to bring up production HOM server at UW today at 1


4/6/2010

On the call: Rob Wynden, Prakash Lakshminarayanan, Mike Klumpenaar, Maggie Massary, Phil Ng, Peter Beninato, Brian E., Hari Rekapalli, Vijay Rayanker, Mark Weiner, Alice L., Deb B.

1) UPenn is doing an install in Mark Weiner's Dev server first then Maggies Production server later

2) Vijay is loading UCare with UETL, Peter is getting feedback on Talend usage in the CTSA, we talked about what little of Talend we still use - just the call to the Bulk Loader for each database vendor - that's it

3) We talked about the ad-hoc assoc of fact table records via HOM

4) We talked about the 1.4 upgrade path - should we upgrade now? - Peter and Brian have looked at it. - it sounds like the Proj Mgmt cell is rough but we should go for it anyways.

5) We talked about Hari's work on the SHRIMP tool

6) Prakash is working with an agent based framework for running HOM maps in the background


3/30/2010

On the call: Rob Wynden, Matvey Palchuk, Aaron Mandel, Prakash Lakshminarayanan, Marco Casale, Vijay Rayanker, Peter Beninato, Aaron Abend, Maggie Massary, Mark Weiner

1) Prakash is done with RxNorm.

2) Assisting with HOM install at UC Davis.  UW seems to be working now.  The update script is done and need to be added to the public site along with the obs fact view update.

3) Vijay updated the UETL feature to use the Modifier_cd (and Encounter) to support logical grouping of records

4) The U Denver workbench is working in Dev but it's still not in Test

5) UPenn just got approved for the wiki where our software will be working

6) Prakash is about to begin work on the implementation of Lab Norm


3/23/2010

On the call: Rob Wynden, Aaron Mandel, Mike Kumpenaar, Prakash Lakshminarayanan, Marco Casale, Tim Thimman, Curtis Hendrickson, Maggie Massary

1) pending - UCSF will deploy MS SQL Server into Test so that we test against all 3 platforms here - Rob will assign this to David Oppenheimer

2) HOM update script under construction by Prakash making installs much easier - will be done today

3) pending - UDenver code working on the Eclipse client - works with HOM already (we are working on an Encounter definition issue on data load)

4) Rochester - 1.4 is running now ... will install HOM on 1.4

5) Prakash to send install instructions to Maggie


On the call: Rob Wynden, Nick Anderson, Justin Prosser, Prakash Lashminarayanan, Vijay Rayanker, Hari Rekapalli, Marco Casale, Phil, Aaron Mandel, Aaron Abend, Matvey Palchuk, Peter Beninato, Mark Weiner, Mike Kumpenaar

3/16/2010

1) UCSF will deploy MS SQL Server into Test so that we test against all 3 platforms here

2) UCSF will ship and an update script for all three platforms that don't assume HOM is installed on a clean i2b2 install

3) UDenver code working on the Eclipse client - works with HOM already (we are working on an Encounter definition issue on data load)

4) RxNorm - we need to implement the obs fact view (The Obs Fact view is not running anywhere besides UCSF, we will work with doing this at UW next)

5) Vijay is working on the large document (over 5 million records) chunking issue - DONE

6) Hari is working on the SHRIMP code.  The code is looking pretty rough.  Not very modular at present and needs a code update.  Recom will help coordinate our branch.

7) http://www.ontomapper.org public web site is up now.  We needed to update the license statement to BSD there before we start using it.

8) SHRIMP UI spec for integration with HOM will be done within a day or two

9) Prakash needs to update the links in the documents.  Aaron will give Prakash access to the public site.

10) Rob will have Dave Oppenheimer working on end user documentation for the HOM app server (end user UI)

11) Rob will prepare a feature roadmap for HOM and put it on the public site

12) Prakash is working a bit on the RxNorm docs so that it explains the connection with UETL better

13) Prakash to start on the R interpreter work

14) Marco asked about 1.4 and query by value.  Peter has just started looking at it.  Rochester will be the guinea pigs and try to get HOM running on 1.4.


3/9/2010

On the call: Rob Wynden, Hari Rekapalli, Marco Casale, Tim Thimman, Brian Eliason, Aaron Abend, Matvey Palchuk, Deb Batson, Peter Beninato, Vijay Rayanker, Prakash Lakshminarayanan

1) Peter has been working on the i2b2 1.4 install.  Looks like he's the first to try.  Not working yet.

2) Prakash has fixed most issues with our current build requests.  He is working on the plug-in architecture and it seems to work as a derivative of the PDO plug-in to i2b2 workbench (AJAX Client).

3) Prakash has figured out that if you use UETL that records don't collapse on input to the database.

4) Prakash will now look at the UDenver code this week.

5) We are releasing a patch for the Obs Fact View as a separate patch for i2b2 client(s).

6) The public web site is going live very soon (later today).  We may need to adjust the documentation to match the new site. Rob will review the site soon. - OntoMapper.org.

7) UETL, Vijay is loading the UCare data.   He is breaking UETL files into chunks automatically.

8) A combined build for HOM including UETL is pending.

9) Hari is looking at SHRIMP docs and build environment.

10) Matvey is still working on the new UI spec for SHRIMP.


3/2/2010

On the call: Rob Wynden, Aaron Mandel, Prakash Lakshminarayanan, Marco Casale, Tim Thimman, Phil, Chandra Bollalbally, Aaron Abend, Peter Beninato, Deb Batson, Brian Eliason, Mark Weiner, Curtis, John (UAB), Nick and Justin, Vijay

1) We chatted about the 1.4 upgrade / merge project which is pending

2) Prakash is working on a plug-in for i2b2 workbench - we may be able to implement all changes as a plug-in (obs fact view mods may be included possibly)

3) Hari just hired - Ruby programming on SHRIMP will now begin (Nick and Rob will talk offline)

4) Vijay is using the UETL to reload UCare.  We can now support builds on UETL at other sites.  Tim Thimman at Rochester would like to give it a try on 1.4.  Tim would like to start today.

5) A combined JAR file will need to be created

6) Public web site for HOM still pending with RecomData.

7) At UW HOM is building the SQL Server version of HOM it seems to be initially working now (based on the Sybase build).  On SQL Server 2005.  Justin will send install instructions back to Prakash to go into the docs.  Elishema will update the docs and help us update them!  Awesome.

8) UPenn is running on a production setup there.  Maggie will implement at UPenn soon (Mark will follow up).

9) Brian at UDenver will send us the source code for the filter (Thursday or Friday)

10) Recom is pretty certain that the plugin architectures are different on the eclipse and ajax clients

11) JSF architecture may help (Justin)

12) Denver is working on eclipse client (JSF will not work).  Is a build in RCP with an export in JSF possible (Brian)?

13) We may be able to use Andy McMurrey's SHRINE architecture to help with the caching of data for the UDenver Query By Value filter (caGRID would be cut out)

14) RecomData may be able to suggest some 1.4 features to help with the server side cache filter port (Aaron Mandel may be able to find some 1.4 docs on this ... server side cohort processing on the server side)

15) RecomData suggests that we may want to consider going with the Eclipse Client!!!!  Then AJAX for cohort selection and Eclipse for Query By Value and let the sites implement the new features on AJAX for now.

16) LabNorm design is ready.  We will use Location Based Normalization (R) and PKDiscretization (Weka).  Waiting for Prakash's completion of the plug-in so that he can focus on this.  Vijay may be able to help.

17) Talked through issue of versioning Talend jobs into SVN


2/23/2010

On the call: Rob Wynden, Aaron Abend, Prakash Lakshminarayanan, Peter Beninato, Tim Thimman, Nick Anderson

1) Prakash is working on the plugin architecture for the i2b2 workbench integration

2) The folks at U Denver are sending us their code today.  We can begin integration with that workbench now

3) Peter: with a javascript wrapper it may be possible to have one plug-in that works with both the AJAX client(s) and the Eclipse client(s). The UI support across the two platforms may be rough

4) The LabNorm design is done. Waiting for Prakash to finish the i2b2 workbench(s) integration first

5) UW is struggling with the SQL Server based install - we talked about some way to address this with DAO improvements.  Not in 1.4?!

6) RTI is in fact moving ahead with integration (Tim and Rob had a call with them)

7) Rochester has HOM running with the RxNorm script (running on Oracle)


2/16/2010

On the call: Rob Wynden, Prakash Lakshminarayanan, Nick Anderson, Justin Prosser, Marco Casale, Tim Thimman, Chandra Bollalbally, Matvey Palchuk, Aaron Mandel, Aaron Abend, Mike Kumpenaar, Mark Weiner

1) Prakash is working on the row collapse feature. (so that query results sets can be based on both 'and' and 'or' ... right now they ALWAYS collapse which is not very good when retrieving records that may not be based on the number of 'patients' but instead may be based on the number of 'records')

2) Rochester is working on the build right now.  Recom to follow soon

3) UW doing HOM install on SQL Server

4) HOM will soon have its own version of workbench (we need to use either the Cincinnati or UDenver as a base system ... or use default Harvard workbench(s)?).  We deferred resolution of this issue until next week following some research and side discussions


2/9/2010

On the call: Rob Wynden, Prakash Lakshminarayanan, Marco Casale, Tim Thimman, Nick Anderson, Justin Prosser, Prakash Lakshminarayanan, Matvey Palchuk, Mike Klumpenaar

1) All sites should rebuild on the most current build of HOM

2) Prakash will post the RxNorm test data for regression

3) We had a demo of the RxNorm script on the call

4) Semantic Branded Drug may be a good additional mapping for RxNorm

5) We may need to add "strength" to the mapped result

6) We allow Drug Form to expand to include children of type Clinical / Branded Drug or Pack (with children restricted to the correct form)

7) There is a flag in i2b2 to allow query using same visit!!

8) We need Davera to give us test data with strengths included (we need this to complete #6 above! Or we can use the test data in #10 below)

9) NDFRT (see if we can get a pre-release from Matvey)

10) There ARE 8,000 test items in the i2b2 test data set that DO include strength data - way may be able to use this!

11) We need a public site for HOM so that we can post our official build now

12) RxNorm V2 will happen later.   We are normalizing labs first.  Also Prakash will result the hardcoded i2b2 "ANY" attribute to support "SAME" with a checkbox to handle synonyms

13) Prakash will also add a tooltip that shows the path(s) where a mapped item comes from


2/2/2010

On the call: Rob Wynden, Vijay Rayanker, Prakash Lakshminarayanan, Marco Casale, Tim Thimman, Ying Qin, Aaron Abend

1) Ontomapper builds now running at RecomData, Rochester and we think OSU (we will confirm that).  Prakash will get UPenn running now.

2) We talked about the RTI component to see if it fits as a new feature for V2 of HOM (An advanced data validation engine that runs based on rules encoded in XML (much like HOM!)

3) Several types of data validation are handled.  Very comprehensive.  All written in java.

4) Data transformation (both a pivot function and a row splitter! and can also do foreign key lookups!)

5) We should be able to get that under BSD for version 1.0 (for version 2.0 there may be some licensing issues).  Version 1 has most of the features.  No transformation in version 1 though.

6) Rob will arrange a call with tech transfer and the RTI lawyer to figure out how to work with RTI.


1/26/2010

On the call: Rob Wynden, Marco Casale, Nick Anderson, Prakash Lakshminarayanan, Chandra Bollalbally, Tim Thimman, Aaron Mandel

1) We need to allow the 'select distinct' in i2b2 workbench to be optional.  Prakash will add a checkbox to the query box

2) Once the workbench is updated then we will work on the 'R' interpreter setup


1/19/2010

On the call: Rob Wynden, Aaron Mandel, Prakash Lakshminarayanan, Matvey Palchuk, Maggie Massary

1) OSU OntoMapper installation... we don't have confirmation but they may be done.  We need to help Aaron now and then move to Maggie and UPenn

2) View mechanism is done and in test

3) RxNorm ontomapper script is checked in

4) Rob is working on the lab data normalization script design


1/12/2010

On the call: Rob Wynden, Kent Anderson, Davera Gabriel, Aaron Mandel, Aaron Abend, Vijay Rayanker, Nick Anderson, Prakash Lakshminarayanan, Justin Prosser

1) UETL Demo (this feature will be merged into the OntoMapper jar

2) SHRINE programming for Mapping Workbench ... hope to be done by tomorrow

3) Prakash is integrating the View with the i2b2 workbench

4) Regression tests for OntoMapper have been written in Test Complete.  UCSF is moving into Production now and we are helping OSU with a build as well.


1/5/2010

 On the call: Rob Wynden, Nick Anderson, Justin Prosser, Prakash Lakshminarayanan, Vijay Rayanker

1) RxNorm status - demo on UETL call tomorrow

2) SHRIMP Work Pending - demo of new UI in 1 week

3) Support for Ontology Mapper on SQL Server at UW - Justin to try using the Sybase IQ code

4) Initial design of Lab Data Normalization script


12/29/09

NO CALL DUE TO THE HOLIDAYS


12/22/09

On the call: Rob Wynden, Prakash Lakshminarayanan, Mark Weiner, Vijay Rayanker

1) We talked about RxNorm.  It's working with Ingredient, Precise Ingredient, Clinical Drug Form and Dose Form.   We ran through some small issues with the XML format and Prakash will work on this over the next week or 2.  Hopefully this will be working around the beginning of Jan.


 12/8/2009

On the call: Rob Wynden, Aaron Abend, Nick Anderson, Vijay Rayanker, Prakash Lakshminarayanan

1. RxNorm web service has been accessed ... it's working.

2. Can find concepts for single word terms only ... is this a restriction in RxNorm?  Hyphens or Spaces throws off the RxNorm API.

3. Prakash is trying to debug multiple word terms now... he may use a different call like GetDrugs for example...

4. The Ontomapper installation stuff is posted for Oracle and Sybase IQ.  Justin at UW may try to get it working with SQL Server also...


On the call: Rob Wynden, Nick Anderson, Vijay Rayanker, Aaron Abend, Prakash Lakshminarayanan, Mark Weiner

1. The Ontomapper UI is done and Prakash is checking in code

2. David O. is starting the test matrix and regression tests (Test Complete)

3. Continuing in on the RxNorm ontomapper scripts


12/1/2009

On the call: Rob Wynden, Vijay Rayanker, Prakash Lakshminarayanan, Aaron Abend, Aaron Mandel, Mark Weiner

1. Dave O. will be ready to start our regression tests soon (he is on vacation)

2. Prakash will send installation docs to Recom by this week

3. Rob is continuing forth on the SHRIMP UI redesign to add a Mapping Workbench Tab for OntoMapper


11/24/2009

On the call - Nick Anderson, Prakash Lakshminarayanan, Aaron Mandel, Mark Weiner, Matvey Palchuk, Mike Klumpenaar
 
1. Nick was interested in progress of RxNorm API for Ontomapper and Prakash reported that this is being initiated at UCSF.
2. Aaron Mandel checked on the status of Ontomapper documentation.  Prakash said that it's is not yet ready and would be taken up shortly.
3. Prakash said that Ontomapper has been tested on HSDB maps.  Aaron queried on the type of medications supported by HSDB. Prakash explained HSDB maps were simple and based on raw HSDB data. Prakash also said that RxNorm maps would be a good benchmark for Ontomapper.
4. We discussed on various RxNorm medications.
5. Rob and team would work on leveraging SHRIMP UI for Ontomapper work.


11/17/2009

 1) There was NO ONTOMAPPER MEETING today because so many people are at the AMIA conference... we reconvene next week.


11/10/2009
On the call: Rob Wynden, Prakash Lakshminarayanan, Davera Gabriel, Vijay Rayanker, Mark Weiner, Aaron Abend, Michael Klumpenaar

1) The reviewed the SHRIMP UI.  So basically the mapping workbench would be merged with SHRIMP.

2) Rob needs a build a build of SHRIMP on my laptop (Ruby on Rails)

3) Rob is prepared a proposed design for version 2 of SHRIMP

4) Prakash walked us through the final XML spec


11/3/2009

On the call: Rob Wynden, Prakash Lakshminarayanan, Matvey Palcheck, Aaron Mandel, Aaron Abend, Mike Klumpenaar, Vijay Rayanker, Mark Weiner

1) Rob Submitted a paper to the AMIA CRI Summit on OntoMapper

2) Recom led a demo of the SHRIMP UI

3) The SHRIMP UI is indeed very close to the Mapping Workbench spec required by OntoMapper.  Rob will take screen shots of SHRIMP and mock up a proposed screen flow for the next version.

4) We will review the new version proposed slides on the next call.  Thanks!


10/20/2009

On the call: Rob Wynden, Aaron Abend, Matvey Palcheck, Nick Anderson, Mike Klumpenaar, Vijay Rayanker Prakash Lakshminarayanan

1) Updates to XML spec will be distributed
2) Workbench allocation - Prakash after the all-hands, UW has no
resources but they do have some contractor money
3) RecomData has verified the current mapping workbench build, still
no word from Aaron Abend on resourcing
4) SHRIMP is written in Ruby on Rails (code is not available just yet
under open source)
5) Do we adapt SHRIMP to output XML files for OntoMapper?  Or do we
just finish the Mapping Workbench?  We don't know.

6) THERE WILL BE NO CALL NEXT WEEK AS WE ARE ALL TRAVELING TO THE CTSI
ALL-HANDS MEETING


10/13/2009

On the call: Rob Wynden, Nick Anderson, Ketty Mobed, Aaron Abend, Mike Klumpenaar, Prakash Lakshminarayanan

1) Review of XML spec with Marcin P at UW.  We need to eval how many resources RecomData can kick in

2) Aaron will look at his resourcing for RecomData and see if they can help on Mapping Workbench

3) We walked through the OntoMapper XML spec

4) As a result of our call we decided to pursue several options all at once:

a) Aaron Abend will evaluate resource allocation at RecomData to see if we can finish the Mapping Workbench

b) As a backup UCSF will just encode calls to the RxNorm directly into an XML script to support instance mapping for CICTR (using Davera's list of 100)

c) We will also evaluate SHRIMP.  Even if OntoMapper ships with the RxNorm calls we may still need SHRIMP unless the hierarchy exposed by RxNorm is user friendly. 


10/6/2009

On the call: Rob Wynden, Calixto Melean, Shannon Hastings, Vijay Rayanker, Nick Anderson, Prakash Lakshminarayanan, Paul Norris, Aaron Abend, Aaron Mandel, Mike Klumpenaar, Matvey Palcheck, Davera Gabriel

1) UCSF just hired a QA resource ... he should start very soon.  There may be some short term resources in

2) We talked about the installation of Globus and caGRID.  There is a graphical online installer available that we should be using. (it's been live for 9 months). Select configure our node to select "Triad" for security and services.

3) OSU is also creating a VM based distribution as well.  (with almost everything configured)

4) Integration of OntoMapper in i2b2 installation package at the code camp at the end of Oct.

5) We need a production deployment for the HSDB (Human Studyome)

6) Installation of caGRID and Triad already includes i2b2 Introduce and Gaards

7) Proposal of a VM distro was a production VM

8) We talked about regression testing and documentation staffing

9) Shannon Hastings will talk to David Ervin about getting a production quality of the new version of Introduce


9/29/2009

On the call: N/A

1) THIS CALL WAS CANCELED. BECAUSE OF A SCHEDULING ERROR THE CON-CALL NUMBER WAS DOUBLE BOOKED.


9/22/2009

On the call: Rob Wynden, Matvey Palcheck, Paul Norris, Prakash Lakshminarayanan, Mike Klumpenaar, Aaron Mandel, Vijay Rayanker, Ketty Mobed, Nick Anderson, Davera Gabriel

1) UCSF Documentation and Testing tasks are on hold until next week.  We are working on it...

2) We talked about instance mapping into the new ICD9 hierarchy.  Prakash got that working with a single level hierarchy but Matvey would prefer that we use the multilevel hierarchy.  UCSF will work on that for next week.


9/15/2009

On the call: Rob Wynden, Matvey Palcheck, Paul Norris, Prakash Lakshminarayanan, Mike Klumpenaar, Aaron Abend, Aaron Mandel

1) Test plan posted. Regression tests is focus week.

2) Installation manual starts this week.


9/8/2009

On the call: Rob Wynden, Aaron Mandel, Aaron Abend, Hillari Allen, Vijay Rayanker, Nick Anderson, Matvey Palcheck, Paul Norris, Davera Gabriel

1) Hillari's test plan is in the completion stage.  She will meet with Prakash later today to get the scripts going.

2) We reviewed the Mapping System comparison chart (checked into the Files section...)


9/1/2009
On the call: Rob Wynden, Hillari Allen, Nick Anderson, Mark Weiner, Paul Norris, Vijay Rayanker, Matvey Palcheck, Mike Klumpenaar, Davera Gabriel, Aaron Abend

1) We should document Ontology Mapper as a system to install as a downloaded VM.  Raw install would come later.

2) Hillari has prepared a test plan.  TestComplete for testing all of it.

3) User manual for Ontology Mapper - Hillari has a requirements that can serve as the basis of a user manual.


8/25/2009

On the call: Rob Wynden, Hillari Allen, Ketty Mobed, Nick Anderson, Marco Casale, Tim Thimman, Aaron Abend, Aaron Mandel, Paul Norris, Mike Klumpenaar, Dave Ervin, Vijay Rayanker, Matvey Palcheck, Davera Gabriel

1) cQL over caGRID query of i2b2 for Ida Sim's HSDB Grid ... it works!  It looks like OntoMapper is feature complete for that sprint goal.  Thanks Everyone!

2) caGRID Portal as a graphical query UI for caGRID ... will be made to work with OpenMDR ... OSU will lead that effort

3) Hillari to work on the XML spec v2 on ctsPedia.org and ratify that spec with Dave Ervin at OSU

4) Mapping Workbench focus.  Shrimp is going open source next week.   We talked about the overlap between these two projects.

5) Rob to create an 8 x 8 grid with features of Ontomapper, Shrimp and Mapping Workbench, Manual Instance Mapping (pre-conversion, proscribed standard, use of OpenMDR, can describe an instance map, association mapping)

6) Harvard would like to allow SHRIMP to handle multiple ontologies.  Harvard may not have dealt with "query by value" yet which is why they are not already working on that issue.

7) Our new focus on this call is now on UW Mapping Workbench, Documentation, Installation Instructions, and Regression Testing (JUnit with Ant(Maven) ... Atlassian Dev Tool Based? ... JUnit now available in i2b2 so that may be a good place to start although the test data was never posted by Harvard)


8/18/2009

On the call: Rob Wynden, Marco Casale, Tim Thimman, Paul Norris, Aaron Mandel, Aaron Abend, David Ervin

1) David demo at 1pm Pacific on Ida's HSDB con call

2) Rob to send out invite for the caGRID demo on Ida's call

3) Focus on docs and testing

4) Integration with SHRIMP on the way (but first we need to complete the caGRID work)

5) Prakash to help Tim with OntoMapper install at Rochester


8/11/2009

On the call: Rob Wynden, Aaron Mandel, David Ervin, Shannon Hastings

1) The end to end test of a query of i2b2 from caGRID works!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

2) Real CQL queries actually work and return live data!!!!!!!!!!!!!!  The HSDB grid is now launched!

3) Our HSDB Sprint Goal is now met - we will now focus on just putting more data into this working grid!!!!!!!!

4) For the HSDB caB2B interface we should have the caGRID data service running at UCSF.  Rob will talk to Rakesh Nagarangan's team and get them started with the data model.

5) Rob will send our HSDB instance maps to WUSTL.  The maps should contain all of the information WUSTL needs to get started.

6) UCSF will work with OSU to get our caGRID data service VM running - Shannon will put us in contact with the people to help us (Justin Permar's team) do that (they may have plans to have a VM preconfigured - Shannon has a CentOS version that we can use unofficially)

7) David has agreed to provide a demo of HSDB for Ida's call next week or so


8/4/2009

On the call: Rob Wynden, Hillari Allen, Nick Anderson, Aaron Abend, Mike Klumpenaar, Vijay Rayanker, Mark Weiner, Marco Casale, Tim Thimman, Alex Koutmos, Davera Gabriel

1) David Ervin is out sick.  Things seem to be fine but we await final confirmation

2) Rakesh would like a demo of Mapping Workbench when it's ready

3) Hillari is awaiting final OntoMapper build today

4) Mark still wants to target JAMIA or J. BioInformatics (JAMIA special issue on Ontologies).  Rob will redistribute the next version of the OntoMapper paper (with Mark's edits)


7/28/2009

On the call: Rob Wynden, Paul Norris, Marco Casale, Nick Anderson, Aaron Abend, Aaron Mandel, Ketty Mobed, Hillari Allen, Prakash Lakshminarayanan, Vijay Rayanker, David Ervin, Joe Adler

1) The HSDB data is now loaded into i2b2, we are ready for query

2) Encoding Project field - should this be moved to a different Encoding table? We determined not... but we do need to use a consistent service URL for caDSR. David will email the service URL to Prakash.

3) We walked through schema with David and Prakash

4) There seems to be an ETL problem with the TrialBank data... Vijay will check into that

5) Prakash has the encoding values assigned to the source instead of the target ... Prakash will fix that today

6) Prakash and Vijay will send a confirmation to David Ervin once we are ready for him


 7/21/2009

On the call: Rob Wynden, Marco Casale, Tim Thimman, Alex Koutmos, Hillari Allen, Vijay Rayanker, Nick Anderson, Mike Klumpenaar, Ketty Mobed, Paul Norris, Prakash Lakshminarayanan, Mark Weiner

1) RecomData and OSU are both presenting at the caBIG conference.  RecomData is providing an Ontology Mapper presentation and their poster is in our files section.  OSU will be presenting the OpenMDR technology.  OSU will not attend the call today for that reason

2) The mappings for HSDB have been created by Prakash.  All that we have left is to get confirmation from OSU that the HSDB has been launched completely on caGRID

3) The team will wait until next weeks confirmation from David Ervin before we close the current HSDB related Sprint Goal.  At that point we will focus solely on documentation and creating a regression test suite for OntoMapper


7/14/2009

On the call: Rob Wynden, Vijay Rayanker, Nick Anderson, Paul Norris, Joe Adler, Hillari Allen, Ketty Mobed, Prakash Laksminarayanan, Alex Koutmos, Marco Casale, Tim Thimman, Aaron Abend, Aaron Mandel, Mike Klumpenaar, David Ervin, Davera Gabriel, Mark Weiner

1) miniOCRe has been annotated ... David asked for XMI ... Rob will send ASAP ... done

2) Prakash is working on the mappings for miniOCRe

3) Simona finished the spreadsheet for miniOCRe on CTSpedia.org.  The spreadsheet is not uploaded yet because Simona uses a Mac... we are updating the docs to include Mac

4) Mark is going to help us get some publication together - our current papers have been rejected and Mark is going to attempt to get something into print for us


7/7/2009

On the call: Rob Wynden, Marco Casale, Vijay Rayanker, Prakash Lakshminarayanon, Paul Norris, Aaron Mandel, Aaron Abend, Mike Klumpenaar, Nick Anderson, Justin Prosser, Bill Grant, David Ervin, Davera Gabriel

1) David looked at domain model and checked miniOCRe in and has an XML schema created for the Introduce Tool

2) CDE should be annoted in a tag-value with a particular name using EA (David will send to Rob who will annotate miniOCRe and resend to Dave)

3) Ida's HSDB data is now loaded into i2b2, concept dimension paths are set up

4) Rob will write 14 instance maps (Prakash will run them).  Those 14 items will form the initial (small but very scalable) launch of HSDB

5) Dave will search each of the 3 obs fact tables from Introduce using the relevant Concept Paths (Prakash will help him to prepare the queries)

6) David had gotten a preliminary end-to-end test going - now he seems to be off critical path

7) We appear to be very close to getting all this actually working with HSDB's live data


6/30/2009

On the call: Rob Wynden, Tim Thimman, Hillari Allen, Alex Koutmos, Ketty Mobed, Nick Anderson, Vijay Rayanker, Justin Prosser, Shannon Hastings, Dave Ervin, Bill Grant, Prakash Lakshminarayanon

1) Vijay has finished loading

2) Domain model and XSD's

3) Rob to email miniOCRe.xsi to David Ervin

4) We can use the OpenMDR plugin for EA to annotate miniOCRe if necessary (we need to add the CDE's for OntoMapper added to the domain model)

5) Dave has a simple prototype working from the OSU side

6) Open in EA and add a property to each attribute with the value of the CDE in OntoMapper

7) Ketty and Hillari will have Simona document her stuff on the CTSpedia site

8) Rob will work with Davera to look up the CDEs

9) Ketty showed us the Logos (2 votes for the bottom one...)

10) Docs are on the CTSpedia site

11) We will review demographics later after CICTR takes that up

12) Hillari showed her documentation for the CTSpedia site so far

13) Version 2 of the OntoMapper UI should be done in 2 weeks


6/23/2009

On the call: Aaron Abend, Aaron Mandel, Mike Klumpenaar, Vijay Rayanker, Nick Anderson, Bill Grant, Hillari Allen, Ketty Mobed, Rob Wynden, Paul Norris, Davera Gabriel, Prakash Lakshminarayanon, Marco Casale, Tim Thimman, Alex Koutmos, Shannon Hastings, Marcin Porwit, Dave Ervin

1) Vijay is importing the HSDB data.  ETL should be done tomorrow

2) Hillari needs Simona the index data descriptions on the CTSpedia.org

3) OSU has had several meetings to evaluate the general purpose nature of the interface... they confirmed that it WILL be general purpose and re-useable on multiple projects

4) OSU needs the data model (EA model) annotated with CDE's

5) Shannon is demo'ing the OpenMDR EA plugin for the 10am HSDB con call to help with #4 above

6) OSU - UI is done and the query engine is ready

7) Dave Ervin says he is pretty close to supporting actual queries

8) Encounter ID/Event is manufactured - RecomData uses that method also

9) Dave is working on the end-to-end test right now.  He has some of the concept path queries

10) Mapping Workbench is check into SVN at Recom

11) Marcin is starting to work on Mapping Workbench now... seems to have what he needs for the OpenMDR

12) Hillari demo of CTSpedia work (Conference Plus sort of bombed ... we will reschedule the demo)


6/16/2009

On the call: Rob Wynden, Vijay Rayanker, Aaron Abend, Aaron Mandel, Ketty Mobed, Dave Ervin, Prakash Lakshminarayanon, Bill Grant, Nick Anderson, Marco Casale, Davera Gabriel, Marcin Porwit, Hillari Allen, Mark Weiner

1) Prakash is incorporating Chandra's UI.

2) Davera has improved our XML spec to include more provenance info ... we are linked as closely to CTS II as we can be at this point

3) Davera needs the printed documentation for the OpenMDR API from OSU / UW will need those same docs for Mapping Workbench

4) Vijay is working on the ETL job (should be in Sybase03 by end of week)

5) Dave is working on the queries for testing connection to grid

6) OCRe needs access to the OpenMDR plug-in to help annonate the Value Domains on the UML in Enterprise Architect

7) Rob will get a simplified 111-79 form of OCRe sent to OSU by weeks end

8) Marco's poster presentation at AMIA went very well.  (Mark Weiner helped out too)

9) Prakash to walk Rochester through OntoMapper install at Rochester (questions will be posted on the Recom Wiki) - will update the install docs for Oracle as we go...

10) Aaron Mandel would like to have the Ontology Mapper documentation on the RecomData Wiki

11) RecomData will try to let Google crawl the "i2b2 Forum"


6/9/2009
On the call: Rob Wynden, Marco Casale, Ketty Mobed, Hillari Allen, Vijay Rayanker, Tim Thimman, Alex Koutmos, Aaron Mandel, Aaron Abend, Mike Klumpenaar, Nick Anderson, Prakash Lakshminarayanon, Bill Grant, Davera Gabriel, Mark Weiner, John Gennari, Dave Ervin, Shannon Hastings

1) i2b2 code camp was called off so there is no plans to have a user group meeting where we can merge our separate i2b2 code bases. UCSF is pushing to have our i2b2 updates merged into the main code-base remotely. Partners will most likely allow merged changes on behalf of Harvard and Partners (a staging merge). In short everyone outside of Harvard/Partners is now using the RecomData SVN for revision control ... that's a big deal.

2) Vijay has started a side project to create a universal ETL process for i2b2.

3) CTSpedia is loaded with iMacro based automation to help make data entry easier.

4) Ida's HSDB data is moving into Sybase03

5) Dave is mostly there on the GUI interface for Introduce.  Database connection pooling (with Apache component) to make query processor fast.

6) Prakash's set of queries are good for Dave to get started.  Good job.

7) Prakash will complete Chandra's upgrade to the Ontology Mapper UI

8) Rochester has some deployment questions that will be posted to the RecomData "i2b2 Forum" area

9) John Deshazo is passing the torch (for Mapping Workbench) over to a new resource (Marcin Porwit)

10) Ketty has offered to create a LOGO for Ontology Mapper

11) OSU has got a first Rev of the OpenMDR API for Mapping Workbench ready now for Nick's Mapping Workbench team. There are 4 components, 1) plug-in for Enterprise Architect for DSR and EVS connections during annotation, 2) Domain Model Generator exports xmi file for DSR, 3) OpenMDR Query Service is the component that Mapping Workbench needs: that is ready too, 4) OSU is still working on the ISO 111-79 database for OSU's DSR in the works but it works with caDSR for now

12) As OSU makes updates to the OpenMDR Query Service they will send those to Rob so that we can update the VM in our Dev environment.  That way John Gennari's team can track with OSU

13) UW will switch to the RecomData SVN server for the Mapping Workbench code ... basically we ALL will at this point (see item #1 above)


6/2/2009
On the call: Rob Wynden, Davera Gabriel, Vijay Rayanker, Gail Harden, Paul Norris, Ketty Mobed, Hillari Allen, Nick Anderson, Bill Grant, Tony Black, Justin Prosser, Jim Piper, Dave Ervin, Prakash Lakshminarayanon, Joe Adler, Marco Casale, Mark Weiner

1) Prakash just extended schema following Rakesh's approval.

2) Will send Dave the sample query statement

3) David has been working on the Introduce tool wizards to use the query

4) OpenMDR API is now ready and Jon DeShazo is starting to work on it

5) Hillari provided a demo of our site on CTSpedia.org which is where our source data elements will be documented and were our HL7 OIDs will be tracked.

6) Also thank you Marco for presenting our poster at AMIA and also we may want to review Phillip Payne's slides from AMIA which are in our Files section...


5/26/2009

On the call: Rob Wynden, Aaron Mandel, Hillari Allen, Prakash Lakshminarayanon, Paul Norris, Mike Klumpenaar, Ketty Mobed, Dave Ervin, Bill Grant, Vijay Rayanker

1) Vijay is loading the V1 HSDB (real data) data into Sybase06

2) We will be able to publish HSDB from Sybase06 even though Dave is using Sybase03 for our end-to-end test

3) Dave is using test data on Sybase03 for the end-to-end testing onto caGRID

4) Prakash will get an i2b2 workbench for Sybase03 or show Dave how to execute the SQL Statements that include the Encoding table

5) Aaron Mandel will check to see if Event is the right way to associate rows within the Obs Fact table.

6) Prakash will verify the 03 and 06 schema's to make sure they are the same... especially the Encode table(s) extensions

7) We need OpenMDR API access for UW team.  They have started development on the Mapping Workbench and they need API access now

8) The CTSpedia work is just about completed - Hillari will demo it on our next call


5/19/2009

On the call: Jon Deshazo, Rob Wynden, Ketty Mobed, Hillari Allen, Aaron Abend, Chandra Lakshminpathy, Paul Norris, David Ervin, Shannon Hastings, Bill Grant, Nick, Tony Black, Justin, Davera Gabriel, Mark Weiner, Marco Casale

1) Sybase03 disk was full, it's fixed now (sorry about that David Ervin was stuck because of that problem)

2) Vijay is loading in Ida's data... just a couple days now

3) The Ontology Mapper changes to i2b2 core will not be merged in on i2b2.org's SVN until June 22nd.  Pre-merge we are using the RecomData SVN server

4) OCRe is ready and Ida's HSDB team needs to convert their OWL model to ISO/111-79

5) Davera says there are tools on caDSR that can convert a OWL model into Common Data Elements ... that tool may be able to help us annotate the terminologies

6) Related to #5: Shannon will provide a document on using Enterprise Architect and will include Value Domain (terminology) annotation (this is the path the HSDB team is taking for now and not #5 above)

7) ctsDSR has been renamed to openMDR (or rather Rakesh Nagaranjan's component is now called ctsDSR and the OSU part is called openMDR ... I think)

8) Chandra has a couple of changes but all of the UI has been updated.  Then he will integrate with Prakash's back-end

9) Marco's poster will be shown at AMIA!

10) Shannon will provide docs on openMDR (ctsDSR) and Rob will post those docs on the wiki


5/12/2009

On the call: Bill Grant, Joe Adler, Nick Anderson, Rob Wynden, Davera Gabriel, Chandra Lakshmipathy, Matt Wyatt (UAB), Hillari Allen, Aaron Mandel, Dave Ervin (OSU), Jon Deshazo (UW)

1) Chandra user interface update - new version with better looking screens.  Changes on 2 screens for Hillari and Ketty.

2) Ida's ETL in play with real data.

3) Dave Ervin has Sybase access and schema.  He is setting up MySQL (NODB) also with i2b2.  Vijay to use his own instance for real data.

4) CTSpedia.org - Hillari setting up a user-access info screen to help with submissions to the CTSpedia (and getting accounts etc).  There will be a general review before go-live.

5) Mapping Workbench may have a operational prototype within a month.  John needs access to ctsDSR Dave will refer Rakesh Dhaval to provide a ctsDSR interface spec.  John D will look at Snoggle also.


4/28/2009

On the call: Rob Wynden, Nick Anderson, Paul Norris, Prakash Lakshminarayanon, Chandra Lakshmipathy, Joe Adler, Hillari Allen, Jon Deshazo, Matt Wyatt (UAB), Ketty Mobed, Shannon Hastings, Rakesh Dhaval., David Ervin, Matvey Palchuk (Harvard), Aaron Mandel, Davera Gabriel

1) Sybase03 will contain the HSDB data - Vijay starting today.

2) David Ervin is making a local instance with that same data at OSU for development.

3) Ida's initial HSDB data is not Human Subjects for now so the first node will run queries from the OSU node.  Then we will switch to the UCSF node (grid services) after development.

4) RCT data model is still not annotated ... annotate all from a single parent.  Shannon will review this to see if they have what they need.

5) CTSpedia.org work - Hillari making a template for data model entry for users (COHRI, HSDB, JOHN D, and G.E. UCARE are ready).

6) ctsDSR is our target as a DSR interface ... John Deshazo to work directly with Rakesh Dhaval on access to an API.

7) The call got interrupted by a double-booked con-call line... Rob has followed up with the other team to make sure that it doesn't happen again.. today's call will be finished at 3pm Pacific on the UW con call line... next weeks call will be on the usual number ... there will be no problem next time...


 4/21/2009

On the call: RobWynden-ChandraLakshmipathy-KettyMobed-PaulNorris (UCSF), NickAnderson-JustinProsser-JimDeLorenzo (UW), AaronAbend-AaronMandel (RecomData), DaveraGabriel (UCD), BillGrant (Nautilus), MatveyPalchuk (Harvard), ShannonHastings-DaveErvin-RakeshDhaval (OSU)

1) Davera is now beginning the process of reviewing our functional requirements and verifying that we have everything covered in our XML spec, the Mapping Workbench and the existing OntoMapper UI.  We reviewed the HL7 CTS II spec on the call.

2) Shannon reviewed OSU status for integration. Requested instance for OSU (UCSF will send a copy of our VM's).  They working on this over the next 2 weeks to get up moving.  Hopefully we can get it all working (without regression testing) by then.

3) Nick reported on John D. progress.  They are working on Davera's use cases and can provide a demo in a bit.

4) We talked about having the Mapping Workbench attempt to explain the delta that changes as a data standard is updated.  As Davera has pointed out when a data standard is updated only a few of the nodes actually change.  By displaying only the delta the Workbench could make the job of defining new mappings much easier.  (often as a data standard is updated if a local site is not using any of the updated fields then the local maps may already be compatible with the new instance map set ... which leads us to Aaron's comment below.)

5) Aaron Abend at Recom had an excellent suggestion to have new mappings that are compatible with the deployment at a particular site automatically alert the local administrator.  We could easily provide that functionality by deploying the instance map XML files into a Linux Repo instead of just using SVN.  (that sort of alert is the main purpose of a Repo).


 4/14/2009

On the call: RobWynden-KettyMobed-ChandraLakmipathy-PaulNorris-PrakashLakshminarayanon (UCSF), AaronMandel-JoeAdler (RecomData), BillGrant (Nautilus), MarkWiener (UPenn), ShannonHastings-RakeshDhaval-DaveErvin (OSU), DaveraGabriel (UCD), MarcoCasale (Rochester), NickAnderson-TonyBlack (UW)

1) Our Sprint Goal of getting the HSDB (Ida Sim's Human Studies Database) live is now again active.  Ida has real data to publish onto caGRID and we are ready for testing with OSU using our test data.

2) The OSU and UCSF (Rob, Chandra, and Prakash) teams will meet tomorrow about connection to the Ontology Mapper Encoding table at 12pm Pacific.

3) Rakesh D and Shannon are still working on the RCT (for HSDB) data model annotation.  We are attempting to figure out if the database platform must still be annotated in the generic data model for HSDB.

4) Chandra is finishing up the Perl scripting support and will soon refocus on the V.2 Ontology Mapper UI

5) Davera will upload functional specifications for Mapping Workbench - Those should be checked into the Wiki.


NOTE: THERE ARE A LOT OF OLDER MINUTES STORED IN THE FILES SECTION... THESE ARE JUST THE RECENT ONES.

  • No labels