Anyone interested in configuring a speech dictation module for OpenEMR?

open forum
1

Open source Help wanted!

OpenEMR is the most popular open source electronic health records and medical practice management solution. Our project administrators have recently put together a new and ambitious project roadmap based on global user need. One of these projects has to do with testing out and configuring a speech dictation package called Simon for medical documentation. Simon is open source and multi platform.

If you are interested in a career in QA and/or technical documentation, this project will be very resume worthy and quite fun! It is also for a great cause.

Reach out to me matthewvita48 (at) gmail (dot) com if you would like to work on this! I am one of the project administrators and would love to talk you through the system and the problem we are trying to solve for.

PS: If you are a developer, sysadmin, clinician, translator, documenter, graphic designer, etc there is always plenty of work to be done so feel free to reach out as well. Our community is very awesome.

2

Also, does the NHS solutions being worked on have a particular solution for this?

3

Hi @MatthewVita,

At present there are a wide range of products in use across the NHS but none are open source. We have two categories of product being used:

  • Electronic dictation management systems - these are in use in almost every NHS Trust, they are relatively basic and simply record the clinician’s voice to MP3, and manage the dictation workflow. They usually allow for ‘typing pool’ style grouping of the dictations, which are sent to secretaries for typing into (usually) an electronic letter format. They don’t convert voice to text automatically, this is done by the secretaries.
    Examples are: BigHand, Olympus, Lexacom. I’ve used all three and they seem massively overpriced for what they are - a place to store MP3 with a bit of metadata.
    As far as I’m aware there are no open source dictation management systems - I looked extensively a few years ago, when a GP practice I was working for wanted such a system but couldn’t afford the expensive proprietary solutions in the market (because of this, as far as I know, they’re still using microcassette tapes…!). If you’re aware of an open source dictation management system I’d like to know about it as we would definitely include it in the NHoS distro.

  • Voice Recognition - these systems, which convert voice to text, are starting to be used in the NHS, but their use is less widespread. They’re more expensive and are technically much more complex. KDE Simon is an example of an open source voice recognition system, I’ve had a little play with it about a year ago and Simon looked promising, although it would need further work to be usable for clinical dictation, which I guess is what @MatthewVita is asking for help with. We would certainly like to be able to include a preconfigured open source voice recognition system in NHoS so maybe there’s opportunity for us to work together on this…
    The existing market for this kind of voice recognition is dominated by Nuance’s Dragon Medical, which is very expensive. I did once have a copy of Dragon Dictation myself and trained it with some medical words, but Dragon Medical comes with a medical vocabulary built in.
    In the NHS I’ve seen these systems used, especially in dictation-heavy specialities such as radiology, but I’m not sure what systems they were using.

Hope that answers your question Matthew.

Marcus

4

@pacharanero,

Thank you for this vast amount of information.

As you alluded to, I am interested in the second category. Specifically, I am interested in using Simon. Do you have any experience with setting it up? While I have been looking for volunteers to help with this work, it appears I’m going to end up running with it myself (the life of an open source maintainer… :slight_smile: ).

Any info/tips on Simon is appreciated!

Of course any and all documentation/configuration will be in the OpenEMR codebase and wiki, so I’ll be able to hand it off to your group so we can share and hopefully help build up the Simon community (the lead maintainer left and the community isn’t very active :confused: )

Thanks,
Matthew

5

I only have the most minimal experience with Simon - just the initial setup on my own machine and some trial voice recognition. I think I just Googled my way around it. I don’t have any specific advice/tips I’m afraid.

I’d love to have the time to work on Simon voice recognition for openEMR but I’m afraid I can’t take on any more pro-bono open source activity - I’m maxed out with other projects and work. Good luck and of course if we do any work on SimonVR for NHoS then we’ll share that and hopefully it will be something you can utilise.

1 Like
6

Thanks for that! The information was very helpful.

-m