COME and WORK WITH US building NHS DATA INFRASTRUCTURE that WORKS
@drchrispaton
Digital health and informatics academic at the University of Oxford and University of Otago - interested in artificial intelligence, human-computer interaction and medical simulation. Editor in Chief, BMJ Digital Health & AI.
COME and WORK WITH US building NHS DATA INFRASTRUCTURE that WORKS
One chart, many stories. Predicted renewable energy deployment vs actual deployment.
BE AWARE: SEALION (YOUNGSTER) This little fella likes to jump out of the bushes at you!
The sealion in question
I love this town.
People don't realise how significant this project is. It solves many of the major technical, ethical and research challenges of using EHR data for research. This approach should be much more widely adopted for making use of all the data we collect (at great expense!) in healthcare.
I used to have this machine as a kid - great computer!
My new, very brief article is about economic growth β highlighting nine countries in Africa where incomes have more than doubled since 1990.
If you've been wondering where I went: my team in Oxford (60 of us!) build beautiful tools for NHS data. We help researchers work on the whole nation's GP records while protecting - provably - everyone's privacy. It's big potatoes!
www.youtube.com/watch?v=GRjR...
"We're going to need clinical health systems and professional societies to get very involved in self-regulation."
FDA Commissioner Robert Califf speaks with JAMA+ AI Editor in Chief Roy Perlis about the challenges of regulating #AI in health care.
If you are using an #AiScribe in healthcare / medicine, is it meeting your expectations?
Answer the quick poll here. @pulseit.bsky.social
#AiMedicalScribes #MedSky #Ai #digitalhealth #healthtech
docs.google.com/forms/d/e/1F...
Ok #Ai nerds. Is that everyone in 2024?π€ Aust politicians just released their final report into the adoption of AI.
222 pg report
15 recommendations - Heavy on regulations
I was going to do a thread of each rec, but itβs 2.40am. I need π΄ Will revist tomoz
www.aph.gov.au/Parliamentar...
We are launching a new MSc in #AI Implementation (Healthcare) programme at the University of Birmingham (UK) in 2025
Those just wanting explanations of the terms used when discussing health-AI, can use: www.owkin.com/a-z-of-ai-in... Owkin commissioned & did the artwork, I wrote all the text. Each concept (e.g., LLM) has a short lay explanation, a detailed technical explanation, & further reading. Enjoy.
True. Also, you don't have to keep consenting patients once they have already consented. I'd worry about automation bias and checking output - better structured summaries and patient access to check the note afterwards would help I think.
I think if you have to consent 3 times (privacy, listening device, algorithm output) and then thoroughly check the output for errors, you start to lose the efficiency gains from AI scribes.
Yes, good to see RCTs in this area now. I think might be feasible to do a trial that uses an ambient AI scribe in a real consultation, then use the summary as a prompt to an LLM to generate a diagnosis. Then compare that to a clinician's diagnosis.
Interesting study but I would argue that this is similar to LLM performance on medical exams rather than real diagnostic reasoning. The cases were presented as written vignettes and LLMs have probably thousands of similar vignettes in the training datasets they can use for pattern recognition.
Mid-2011 was when the awesome Grahame Grieve came up with #FHIR. He has changed our world! What a #rockstar!! π #Sparked
The course is a well-known online introductory informatics course. The LLMs scored better than as many three-fourths of students taking the course. The big questions for me are their implications for student learning and assessment.
The original paper is at: doi.org/10.1038/s417...
I realize I been spending so much time choosing others to follow and spreading the word about Bluesky that I have not yet posted any science. So let me begin.
I recently published a paper about the performance of LLMs in a course I teach: informaticsprofessor.blogspot.com/2024/11/gene...
Please add me. Thanks!
Hi Larry! Nice community of digital health folk building on here. :-)
What do you think about doctors using AI scribes? They record the consultation and then use an LLM such as GPT-4 or LLama to generate a summary. Is it safe for doctors to use LLM output for medical notes? We discuss this in our recent seminar: www.youtube.com/watch?v=d7Z6...