Concall Scientific Discourse Task Force October 31 2011

Agenda: tentative use cases as per 31 Oct 2011

1. Enhance Drug-mechanism Knowledge in Drug Product Labels

Leads: Richard Boyce, Jodi Schneider, Anita de Waard
Partners: DERI, Elsevier – still need pharma partner

Also based on:

Annotating Truth Values: Helping Knowledge Discovery Through Hypothesis-Based Epistemic Markup

2. Defining minimal information requirements for biomedical investigations defined and mapped to RDF

Leads: Tim Clark, Susanne Santone, David Shotton

3. Mining Treatment Outcomes: Develop a generic, anonymised, multi-EHS compliant format and an information architecture that allows access to symptom/treatment data as a Linked Data source that can be made freely available and used for drug efficacy/outcome (meta)studies

Leads: Joanne Luciano, Anita de Waard
Partners: RPI, Elsevier, - need clinical/patient forum and/or pharma partner

Notes:

10:08 Anita Richard Boyce will lead use case 1. Status: he spoke with managing editor of ePocrates, a platform for drug information software, free handheld tool - are looking into partnering with use case.
10:10 Anita Writing of UC 1: Happy to follow suggestions on Google doc and get confirmation from everyone on the UC - Mike Taylor from Elsevier combined two UCses (with Hypotheses/Evidence one) into one.
10:13 Anita Rich will lead the rewrite of UC 1.
10:14 Anita Pieder Caduff from Elsevier has indicated we can contribute a number of drug records from the Elsevier drug database Reaxys.
10:15 Anita Rich will select a specific set of drug classes, e.g. antidepressants, often used in nursing home environment to increase appetite, sleeping; investigate drug interactions in regimes in elderly patients.
10:16 boycer please type his name
10:17 Anita MIke - ideas about technology? Epistemic support - have been working on NLP algorithm for PI but need to set up a pipeline for this.
10:20 Anita Mike Taylor: we used GATE to tag textbooks using our internal ontologies; lacked Machine Learning techniques to see if we can work on identifying sentences with the right pattern.
10:20 boycer I2B2 challenge
10:21 boycer drug information extraction from scientific text
10:21 Anita https://www.i2b2.org/NLP/Coreference/Call.php
10:21 Tim http://code.google.com/p/annotation-ontology/
10:22 boycer https://www.i2b2.org/NLP/Medication/
10:24 boycer J Am Med Inform Assoc. 2010
10:24 boycer september
10:25 Anita Could add NLP smarts from ISI/Hovy
10:25 Anita Could use the Harvard Annotation Framework for hosting data and running the pipeline
10:26 Anita We will have a separate call on this UC.
10:27 Tim DOMEO annotation toolkit / framework - webinar http://www.bioontology.org/annotation-ontology
10:32 Anita Use case 2: discussing different projects with IsATab with David Shotton and Susanna Sansone, who have slightly conflicting models - may pick one of the levels of generality, that will work for everyone.
10:34 Anita Better title: Defining core metadata for describing biomedical investigations
10:34 Anita 3. Mining Treatment Outcomes: Develop a generic, anonymised, multi-EHS compliant format and an information architecture that allows access to symptom/treatment data as a Linked Data source that can be made freely available and used for drug efficacy/outcome (meta)studies
10:35 Anita SymTrends - self-tracking symptoms by patients
10:38 Anita Other use cases to the people on the call? Answer: no!
10:39 Anita Everybody still there? Yes! AOB? No. Close the call!