Open Healthcare Framework

Thursday, August 17, 2006

OHF at LinuxWorld (and pictures)

The LinuxWorld Expo Healthcare day had a good OHF attendance. We had representations in three panels and had two demo tables. We had conversations with many other open source healthcare groups, and thanks to Melih we even have some pictures :-). .

Tuesday, August 15, 2006

Just what is OHF then?

I received this in my email:

> I am very much interested in learning about OHF
> and what how I can use it to build applications in Healthcare Industry.
> It seems to be reasonably difficult to find the Architectural Overview or
> a list of High Level Use-Cases supported by OHF while going through
> blogs and wiki.
> Would you recommend a decent place to start learning about OHF ?

It's hard to know what to say in response to this - the web site, the wiki and the blogs are all that there is. Other than the code, of course, but reading code is not going to help with this question.

I know that we should have more and better documentation, particularly about the overall project, it's aims and architecture. One of the problems is that the project is currently a collection of somewhat disconnected pieces of work that do not have an overall coherence and focus. In the plan for this year, we have been very explicit about this, and why. The reason is that no clear architecture leaps out. OHF has a strong server side focus, and this is an area that Eclipse is still growing into. There will be further discussions on this at our meeting in Boca Raton on September 9.

As far as use cases, I don't think we can be so specific. We have the following components:
  • HL7 V2 runtimes and conformance profile editor
  • IHE Actor implementations
  • HL7 CTS Implementation
  • Public Health modelling platform.
What's coming is HL7 V3 and openEHR tooling. This is just a grab bag of useful interoperability support things. If you need to do interoperability using these standards, these components will be useful.

As the OHF project develops and starts delivering real value in the areas it has started with, I hope that we can have a wider scope that will make architectural overviews and use cases useful for describing our work.

Friday, August 11, 2006

Warming engines towords IHE Connectathon 2007

While the electronic registration for Connectathon open date is getting closer, we are focusing on what IHE actors and workflows will OHF support using the IHE Client plugins and Bridge (for LAMP/.NET apps).
While OHF is starting to receive feedback from costumers which are going to use it in the connectathon, additional feedback is welcome.

Tuesday, August 08, 2006

OHF at Linux World Conference & Expo Healthcare Day

Next week we'll discuss and demo OHF at the first ever Healthcare Day at Linux World Conference & Expo. After demonstrating this in the IHE workshop few weeks ago, it will be interesting to hear the responses from a conference that is not IHE focused.
In the panels, Don Jorgenson and James Kaufman will talk about all aspects of OHF. Eclipse OHF will demo in two booths. One booth will demo OHF security, encryption, and digital signature. The other will demo the OHF IHE actors capable of searching, sending, and retrieving Healthcare documents using the OHF HL7 plugins. The demo will use the OHF Bridge to show how different EMR applications interoperate and send documents from one to another, a mini-connectathon :-)

We will be happy to see you there.

Tuesday, August 01, 2006

OHF as an API for Healthcare Interoperability

Among others, the Eclipse OHF project is being used in several events sponsored by Integrating the Health Enterprise (IHE) and Health Level 7 (HL7). The OHF IHE plugins build atop of other core OHF components (security, and HL7 stack for example). OHF members are active in the IHE committees, and share the goal of true interoperability in healthcare.

One could argue that the IHE specifications can be viewed as a distributed "operating system" for healthcare interoperability. Some of the reasons MS Windows™ became a popular "user experience" OS in the nineties is:
  • For developers, Windows provided tools to
    • Write pixels to the screen and use assembly if they really wanted to
    • Use standard components in C++ to make an application more conveniently with windows native UI.
    • Use Visual Basic to make it possible for the novice programmer to create a decent application.
  • For the users they provided a good plug and play experience. It wasn't perfect but the drivers mostly worked, and users where not expected to ever know about source code compilation.
Please don't flame about using Windows as an example. I use and love both Ubuntu and MacOS, just want to make a point of how to help users and ISVs use a system.

How can OHF help make the IHE model the dominant healthcare interoperability architecture? Obviously open standards and technical excellence are not enough. Here is the OHF analog to the list above:
  • For IHE developers, OHF provides tools to
    • Implement the IHE protocols, going down to the IHE "assembly"; i.e. implementing the ebXML, SOAP, HL7 stack and so forth. No need to use OHF for that, though one may use at least few of the OHF building blocks like the HL7 and security plugins.
    • Use the OHF Plugins (such as the provided actor APIs, data structures, and security features) as building blocks to construct IHE compliant applications.
    • Use the OHF Bridge web services interface by existing applications. The bridge exposes radically simplified API to make it possible for the novice IHE user to easily create healthcare interoperability workflows.
  • As we found in the last IHE Connectathon, protocols and specs are great but in real life, implementations seem to create "dialects" of the protocol. OHF users will enjoy a plug and play experience. It won't be perfect but OHF will provide components well tested against vendor services as soon as those become available. Users are not expected to know about ebXML, namespacing problems, carriage return charecters in a SOAP attachment, or the HL7 versions.
So why open source? As the saying in the HIT for small & medium clinics goes "free is not cheap enough", license is only part of the TCO and it's not what open source is all about. The main reason for Open Source is it's collaboration model. Open source provides a great collaboration model between partners and rivals in the IT market. A good example is the Eclipse organization itself where you see a set of partners and rivals BEA, Borland, CA, IBM, Intel, Motorola and Nokia collaborating and sharing actual source code.