HCLSIG/PharmaOntology/Meetings/2009-10-22 Conference Call

From W3C Wiki

Conference Details

  • Date of Call: Thursday October 22 2009
  • Time of Call: 11:00am - 1pm ET
  • Dial-In #: +1.617.761.6200 (Cambridge, MA)
  • Dial-In #: +33.4.89.06.34.99 (Nice, France)
  • Dial-In #: +44.117.370.6152 (Bristol, UK)
  • Participant Access Code: 42572 ("HCLS2").
  • IRC Channel: irc.w3.org port 6665 channel #HCLS2 (see W3C IRC page for details, or see Web IRC)
  • Mibbit instructions: go to http://www.mibbit.com/chat and click the server link. Enter irc.w3.org:6665 into that box, enter a nickname, and enter #HCLS2 for the channel
  • Duration: 2h
  • Convener: Susie

Agenda

  • Ontology Update - Colin
  • Data Conversion - Anja, Trish, Susie
  • Data Loading - Michel
  • Mapping Data to Ontology - Susie
  • User Interface Update - All
  • Use Case Questions - All
  • Paper - Colin
  • AOB

Minutes

Attendees: Michel, Jun, Julia, Matthias, Elgar, Joanne, Bosse, EricP, Chris, Susie

Apologies: Christi, Colin

<ericP> scribenic: michel

<michel> susie: update on ontology

<michel> susie: data conversion

<susie> Susie: loading data

<michel> susie: mapping of data to ontology

<susie> susie: interface

<michel> susie: paper

<michel> susie: update on ontology - colin provided notes

<Susie> Colin's notes: Ontology update: namespace, thanks Susie; useful suggestions from Elgar that I haven't actioned yet; you are all free to make the necessary changes to the ontology based on the mapping exercise and I will adjust ontology OR say what you need and I will tweak.

<michel> susie: w3c namespace : http://www.w3.org/2001/sw/hcls/ns/transmed/tmo

<michel> susie: data conversion

<michel> susie: anya -> mapping of medicare D listed drugs into RDF, from HTML, linked with drugbank & dailymed, will have data for loading soon

<michel> susie: on medicare website seems that a key used, brand in uppercase, generic in lowercase, over the counter light grey colour

<michel> susie: can be eligible for medicare, but through some plan?

<michel> susie: medicare has over the counter, so should be in medicare plan; but people on medicare might also be on insured plan, but don't thave access to over the counter drugs - so why is medicare offering these?

<michel> susie: in use case, none of our patients should be taking the over the counter.... messy and confusing

<michel> joanne: might be a problem in delivery of health care?

<michel> susie: could be discussed

<michel> susie: trish working on diagnostic info available - what terms used in diagnostic information, mapping to existing ontologies

<michel> susie: data sources - some commercial - chris identified a paper with some diagnostic criteria, public data sources are available

<michel> susie: can use bioportal to do the markup

<michel> joanne: todo

<michel> susie: created one patient using indivo PHCR

<michel> susie: contact, medication currently on, lifestyle, pre-existing conditions, immunization, genetic information (family)

<michel> susie: generated physician encounter - getting a medical - biochemistry, hematology, blood pressure, weight, height, etc

<michel> susie: encounter - forgetfulness; sees various experts (cognition test, MRI)

<michel> susie: hopes that Peter Kos to get back with more encounters

<michel> susie: once made, need to convert to RDF and load into triple store

<michel> susie: how much genetic information / drug interactions should be included in the PCHR?

<michel> joanne: google health?

<michel> susie: setup account, but haven't investigated it further.

<michel> joanne: patient-centric, providers can be given access

<michel> susie: still in development with health care partners; lifestyle companies getting involved

<michel> susie: the PHCR schema is useful because it has diverse, requisite and useful fields

<susie> Susie: loading data

<michel> susie: progress with triple store?

<michel> michel: still waiting for more data

<michel> send me the PCHR data when it comes in

<michel> eric: conversion of xml to rdf

<michel> susie: based on indiva schema

<michel> eric: grddl could be used

<michel> i'll just write a quick and dirty script

<michel> susie: mapping of data to ontology

<michel> will query for the types found in the data, can generate the mappings from the list obtained, should be straightforward

<michel> susie: TMO interface design call

<michel> susie: need to think about the questions that people want answers to; begin with basic mockup; get feedback on useability

<michel> susie: multiple interfaces could be designed : physician, researcher, patient

<michel> susie: having a polished interface not realistic for deadline; pubby for now, paper should include some text about the future of the interface

<Susie> http://esw.w3.org/topic/HCLSIG/PharmaOntology/Interface

<michel> susie: tabs to give views, each catered to the topic

<michel> susie: started mockup in ppt file on wiki page

<michel> eric: ODM? has lots of forms

<michel> joanne: categorical forms restrict data entry, which is good, but also bad because it prevent information from being captured

<michel> susie: would be good to have event drivers that can pull information about some entity (right click behaviour)

<michel> ericP: HTML + JS?

<michel> susie: depends on what's available - promising

<michel> susie: what's the user going to do; have been ignoring data entry; browsing the data; querying the data; visualization of result

<michel> chris: should have profiles for each of the roles that we've identified

<michel> susie: we'll focus on interface after the paper

<michel> susie: need to think about what an interface should look like for a physician

<michel> chris: avoid reinventing the wheel - US veterans hospital association - all electronic & clinical, simple, graphical, tracked history, easy to use

<michel> ericp: might be able to get a demo/view at VA system

<michel> bosse: worth pursuing

<michel> chris: extra features brought in by this group will enrich the interface

<michel> chris: pathways, metabolism, biomarkers, treatments and drug responses - could answer questions about the relationship between disease and genetics and drug response

<michel> susie: home page, data entry fields, if there is anything about genetics, responses

<ericP> action ericP to ask Liz V to grab: home page enter patient info patient summary adverse events genetics blood tests

<ericP> action: ericP to ask Liz V to grab: home page enter patient info patient summary adverse events genetics blood tests

<michel> chris: how do their systems displays diagnostic results

<ericP> e.g. graph of blood changes

<michel> lots of interest to discuss this further at the F2F

<michel> Susie: the goal of the ontology is to integrate genomics, side effects, clinical info -> how much should be contained in the PCHR?

<michel> ericP: if the data is not stable, then makinga copy would make sense

<michel> chris: patient records should contain their specific information

<michel> chris: but in this new system, the knowledge is integrated

<michel> chris: user can include relevant info

<michel> Susie: paper - needs contributions from medical people

<michel> Susie: everybody should look at the paper and contribute/edit

<michel> Susie: make sure the messages are put forward

<michel> Susie: extension - no confirmation yet from Trish; will write Barry

<michel> Susie: asks chris medicare d question

<michel> chris: pharmacist enters all the info, submit electronically to insurer, which says covered/not-covered & deductibles; different plans, including medicaid, can allow double billing - two insurers will cover the whole amount. medicare d will double with various carriers, over the counter is plan exclusion, but one or the other might cover it

<ElgarPichler> http://code.google.com/p/translationalmedicineontology/

<michel> action: everybody to look at paper; add comments into the text; if adding text use a colour - legend at the top

Retrieved from "https://www.w3.org/wiki/index.php?title=HCLSIG/PharmaOntology/Meetings/2009-10-22_Conference_Call&oldid=25870"