The increasing ability to electronically maintain healthcare data carries with it the enormous potential to also improve overall healthcare. Collecting and analyzing data opens possibilities to further share knowledge, improve patient outcomes, optimize coordinated care processes and support evidence-based medicine. This is accomplished by comparing the effectiveness of treatments and detecting treatment trends. A data-driven solution is a key component in providing this necessary knowledge for better care, lower cost and improved efficiency. MGRID offers such a solution.
For the AXLE project, Portavita and MGRID have joined forces to create a synthetic healthcare dataset generator. The dataset consists of a HL7v3 RIM database, based on the 2.0 release of the MGRID Healthcare Data Model (HDM) and Healthcare Datatype Library (HDL). To populate this database, an unlimited amount of 47 different kinds of FHIR and HL7v3 Clinical Document Architecture (CDA) XML messages can be generated. A parallel messaging infrastructure takes care of fast transformation of XML persistence to database persistence.
* Healthcare Datatype Library 2.0 * Healthcare Data Models 2.0 * Messaging SDK 2.0
The talk Privacy-aware analytics on healthcare data (pdf, 300KB) was presented at the January 2014 HL7 Workgroup Meetings. It triggered some interesting debates from a legal perspective, such as what to do when there are contradictions in the law. Also there was interest in what the next steps in this project are. This work is funded by the EU FP7 AXLE project, that runs for two more years. The goal of AXLE is to advance large-scale complex analytics on real-world datasets, while addressing the full requirements of real datasets, such as data quality, privacy, security and auditability.
On 2013-11-28 in the HL7 AID UGM meeting we presented "RIM for FHIR".
Abstract: "At MGRID we want to reason about medical data from many sources in a temporal context. The source data can be in many different standard formats (CDAs, v2, v3, and now also FHIR). Our approach is to bring all different formats to the industry common model (the HL7 reference information model), and use analysis to go from message fragments to patient information.
What do you do when you're asked if you can give a short talk about your current work by the co-chair of the RIMBAA group, in one of the next sessions? Say yes now, solve problem 'need to make a slide deck' later.
We presented a summary of work so far, and gave a demo of an example mobile health vital sign measurement. A PDF of the slide deck is available from the document page of the AID (formerly RIMBAA) working group.
Last year we have been busy improving earlier work, on using a Model Driven Architecture approach to generate a RIM database and message parsers (loaders). We have summarized this work in Mastering Medical Data with MGRID SQL (pdf, 3MB), which is built upon datatype work described in MGRID Healthcare Datatypes (pdf, 4MB).
Through our sister company Portavita we are from now on involved in the EU FP7 COMMODITY12 project.
The aim of COMMODITY12 is to improve the daily care of diabetes and prevent/manage its comorbidities. It will improve health workers' interpretation of the patient medical status and support coordination of the care. Furthermore, the emergence care process will improved by an early warning system, reducing the hospitalization rate of diabetes patients.
We presented our latest addition to the MGRID platform at the november 2011 HL7 RIMBAA international meeting. Our talk about persisting HL7 CDA documents was preceded by some interesting talks about a RIM-based application in Finland and a Drug Information system in Canada.
These two talks clearly demonstrated a number of common themes for RIM-based applications: