Augnito/ Scribetech

Disapointingly, having said I hadn’t had the microphone disconnect mid report, it has happened to me twice this afternoon. I’m really hoping this is only a ‘one off’

Hi Frank, have you tried DMO - Dragon Medical One, we tested this a while ago and it was quite impressive, it just worked well from out the box no learning whatsoever and the accuracy was very good indeed. Not tested it with our RIS/PACS but we are using SpeechMagic at present and it freezes, misses words and not as accurate as DMO.

1 Like

My understanding is the Dragon is end of life and no longer supported. Plus it doesn’t play nicely with CRIS.

Hi, Dragon V13 / V15 is end of life as an en site / datacentre deployed instance.

DMO / Dragon Medical One is the new cloud based solution, it appears to be supported by several vendors such as Soliton / CHC.

I assume that Nuance will continue to push DMO and pretend the rest doesnt exist any longer for those of us stuck with it.

But should that not be a problem for the supplier of your RIS to solve?

3 Likes

Just an update re Augnito. We’ve upgraded to CRIS 2.12.12 and are currently trialing Augnito with a limited number of users, due to increase significantly over the coming weeks. Will update as we know more regarding performance, but first signs are positive.

Interesting. So is bug which causes the microphone to turn off after 10 mins fixed? Do you think it is more responsive, and can you turn the dictation box off? Keen to hear more when you have it.

Has anyone had any experience with Recognosco using Magentus (nee Wellbeing) CRIS?
Thanks

It appears to be ok so far, and you can toggle off the dictation box. It’s much smaller and less intrusive but I’ve still not yet encountered a radiologist who wishes to use it. Accuracy is good, only a few things we’ve noticed thus far, and it doesn’t require any training. Templates are easy to set up, or paste across from Dragon. It does take a little more time to populate text, I guess to be expected as it’s Cloud based, and it takes a different way of working, waiting for the text to populate before toggling off the mic. Using a headset works well and reduces the likelihood of forgetting not to toggle off too soon. It’s definitely a big improvement compared with the version we tried back in September on 2.12.06. Will keep you updated :blush:

Thanks

Thanks for the update Andrea. Keep posting please.
The thing that puzzles me though, it that you say you’ve not encountered a radiologist who wishes to use it? Did I interpret that correctly? Who is your test audience? And what is their fall back VR?
We are still stuck on unstable 3-reports-lost-per-day-CRIS-Dragon-15. I need to persuade our radiologists to be willing to try CRIS-Augnito out once more. They’ve lost all faith after 2.5 years.

Some of our radiologists report for Nuffield (Shrewsbury). When will they be able to try this out?

We have been using Augnito with CRIS and Insignia PACS for the last 2 weeks at a Circle Hospital.

I have been really impressed with the quality of the VR. Have dictated whole MR and CT reports without having to make any corrections at all. Have had none of the problems others have mentioned.

It does take a little longer for the text to appear but that is no problem.

Excellent piece of software vastly better than Dragon or any of the other Nuance clones that I’ve had to use in the past.

1 Like

They just don’t want to use the preview dialog box, not the VR itself :slight_smile:

Apologies, realised I only answered one part of your question! Yes to Shrewsbury Nuff radiologists, will check our rollout schedule as we’re doing a few sites at a time, due to specific issues we encountered last time around in September.

Hi All,
We in SWASH have recently upgraded CRIS to 2.12.03c and are now talking to Magentus with a view to testing Augnito.
Just wondered how people were getting on with Augnito that have already started on that journey?

Would be really useful to go into this with eyes wide open!

1 Like

Hi - Just wondered if anyone had any experience or feedback of Augnito they could share?

My experience at a national private hospital:

  1. Pretty good accuracy, but…
  2. No option to train any (repeated) mistakes
  3. Noticeable delay in cloud recognition
  4. Can define “canned text”, but often poor recognition of trigger phrase
  5. No ability to centrally define text across all users
  6. Cannot add custom or missing words
  7. No scripting functionality to move between report sections in CRIS

Hi. Radiologist here.

Since the CRIS upgrade 2.12.12 Augnito integration has been upgraded. Before this it was really buggy and not really usable. Now it is much better.

  • Accuracy is good (probably equivalent to dragon, not as good as Mmodal in my experience).

  • There is a lag between dictating and seeing text on screen - quite a bit longer than dragon as the voice is processed online. There is a bug associated with this:

  • If you turn the microphone off before the text is transcribed onto the screen it has an error where the transcribed text is not outputted properly, and getting rid of this text can be difficult/impossible, and you often end up having to come out of the report and start again. If you share an office or get many interruptions it can be really difficult to turn the microphone off at the right time, as you risk it erroring whilst its trying to process their speech.

  • The microphone turns off after every report which is annoying when you are trying to get through plain films quickly.

Overall with the recent CRIS upgrade I think its at a stage where its usable, and probably less issues with roaming profiles getting corrupted that we had with dragon. But when dragon worked it was better - more responsive transcribing, and no problem turning microphone off during speech.

James

Hi,
Radiologist here who has used Augnito at Nuffield for some months and trying to implement Augnito at local NHS Trust.
Agree with all that has been said. Augnito works fine at the Nuffield. No major issues apart from what has been said in this thread.

At out Trust, we started deploying Augnito end of October and it worked great for a week. Then every single user started having issues with microphone not responding, ie stays red when trying to report and then crashing CRIS. We have moved back to dragon as it was unworkable. IT have made several changes whitelisting and modifying the smooth wall and we have tried moving back to Augnito but the issues persist. Magentus have taken ages to come back with any sort of help and it’s now the end of November. The issue still hasn’t resolved.
So until this microphone issue and crashing issue is sorted, we cannot roll it out to all the Radiologists. So back to Dragon which works well and is reliable given the state of Augnito.
Used to have really good response time from Wellbeing in the past but not impressed with the Magentus support.

We were to upgrade 2.12.07 to 2.12.12 this morning but the Trust cancelled earlier in the week. Magentus had shared the release notes with us (of the usual quality). Three days before the upgrade we had a meeting to discuss “known issues”. There is apparently an document that they had not shared with us in a technical language that even the Magentus engineer could not decipher.
They then announced that some radiologists in one Trust had stopped using CRIS - Augnito but they did not know the reason.
We pulled the plug.

But that leaves us with the unstable CRIS - Dragon 15, 6 frustrated radiologists and 2 radiologists on digital dictation with better blood pressure readings.

1 Like