Cracking the Code on Patient Medical Record Search

Informatics for Integrating Biology & the Bedside (otherwise known as i2b2) is an NIH-funded initiative managed by the Partners HealthCare System here in Boston.  Over the past few months I’ve been spending time with a team of folks that is involved in various parts of this initiative helping them think about new commercial initiatives evolving from i2b2.  It really is an amazing world…hanging out with MD / PhD types and wandering the halls of a medical research hospital is a far cry from a typical tech startup.

Medical-records The background on i2b2 is that a few years ago the NIH invested something like $20 million into developing an open sourced platform that solves a major problem faced by hospitals (and medical researchers) which is that existing hospital information systems (“HIS”) were not designed for search.  A typical HIS is great for finding information about a particular patient.  Type in medical record number 0123456 or a name and, boom, you have every bit of information from profile data, hospital visits, lab data, radiology, billing, etc.  But if you were to instead try to search for patients that were male, aged 37, Asian ancestry, under 6 feet tall and have a particular condition, the system would be unable to answer (actually you’re unable to ask).  So that’s what i2b2 was designed to solve and it’s pretty good at it (you can play around with it here).

This is the real Google for medical records.

The technical challenge is daunting in and of itself, but that’s just the tip of the iceberg.  When dealing with medical records there are the legal issues of patients’ right to privacy so having an ability to de-identify or anonymize data is key.  And add on that a layer of bureaucracy almost unknown anywhere else and you’ve got yourself a hard problem to solve.

Apparently there are now 30-odd hospitals live in some fashion with i2b2 and later this fall (targeted for September) i2b2 v1.4 is slated for release and this version is dubbed the “enterprise ready” version. Program director Zak Kohane describes the next release as:

This release allows the entire set of patients from the enterprise to be exposed to the research community of the hospital while preserving patient privacy, providing the way for basic queries to funnel down to a set of patients interesting to the investigator. The chosen set of patients is matched to controls from the enterprise database and a project-specific mini-database is created. Major software enhancements of release 1.4 compared to release 1.3 include a hardening of the project management cell with a new, full featured client for managing enterprise users, a plug-in architecture for the data repository cell that allows complex server-side workflows, and streamlined project mini-database creation with processes to build sets of matched control patients for a given set of patients of interest.

Now there are many for profit companies also attempting to solve this problem including Microsoft, IBM and others but I think open source is a credible competitor in this sector as it has been in so many others.

So this coupled with a huge influx of public and private capital invested in “online medical records” is going to create some very interesting opportunities.  Hopefully I’ll be able to talk more about this particular project soon.  In the meantime, if this is an area of interest or expertise for you, I’d love to hear from you.