Shared Health Research Information Network is the i2b2 framework to allow federated queries across multiple healthcare systems that is now being adopted by several national efforts, even while we continue to define it.
Tuesday, May 26, 2009
Friday, May 22, 2009
Rheumatoid arthritis
Friday, May 15, 2009
Rheumatoid Arthritis and Smoking
Plenge et al.,
Family history mentions smoking and that seems to create some false positives. Also the negex negation algorithm missed a few negations. Discussed alternative approaches elaborated in the i2b2 NLP smoking bakeoff.
Also reviewed cardiovascular risk in the medical record and the gross lack of specificity of myocardial infarction obtained from claims/billing data as compared to the narrative text.
Storage
Richter, Murphy, Keogh, Kohane
Reviewed the very significant infrastructural requirements for i2b2 instances and the multiple data marts that we are deploying. Brent has two interesting solutions at HP and Sun that we are exploring.He will report back in one month with the results.
Authentication
Murphy, McMurry, Kohane, Weber, McGow
Discussed some of the challenges of different namespaces for authentication across different i2b2 instances. All the more challenging with the proliferation of SHRINE implementations. Discussed different national solutions such as Shibboleth.
Friday, May 8, 2009
Rheumatoid Arthritis
Plenge et al,
Reviewed upcoming i2b2-related grants.
Sergey Goryachev reported on the latest NLP task.
Reviewed CCP positivity vs erosions in i2b2 vs a prospectively accumulated registry (i.e. BRASS)
Reviewed year 6 research questions.
Major Depressive Disorder, Resistant to SSRI
Smoller, Iosifescu, Perlis, et al.
Introduced Russell Hanson
Reviewed the pharmaco-vigilance component. Discussed ways to discretize time for exposures.
Tuesday, May 5, 2009
i2b2 Academics Users Group Meeting
Because of fears of the swine flu, most of the Boston i2b2 team decided not to fly to Seattle. Instead NIck Anderson graciously allowed us to hold the meeting by web meeting technology.
Nice write up of the meeting by Connected-Science
Friday, May 1, 2009
Rheumatoid Arthritis
Plenge et al,
With NLP task pulled off, the trigger was pulled on the samples which are now starting to flow in.
Details of control mart.
Discussed co-occurence of CCP positivity and erosions and the disparity between NLP-driven results and some of the published literature.
Denominator issues dogging us again.