Machines that understand medical terms in Norwegian
Watch the recording about how to get artificial intelligence to evaluate clinical texts and what other tasks they can accomplish.
Spoiler alert! It worked for us too. Now we have an improved model for understanding medical texts.
Imagine if artificial intelligence could read the discharge summary for you and assign the correct diagnosis code. Or that it could process a referral and suggest a reply.
In order to make this happen, we need artificial intelligence that understands medical terminology and the language the patient and doctor are speaking.
Presentation by Christian Autenried, head of Helse Vest IKT's artifical intelligence team.
Helse Vest IKT is working to further develop language models to understand medical terms in addition to Norwegian. During the webinar, Autenried told us about how they train the models, what the models will be able to do to support healthcare professionals, and the spread of the models in Helse Vest and to other regions. He touched on topics such as validation, synthetic data and privacy.
Christian Autenried has a degree in mathematics from Ludwig-Maximilians-Universität München and has a Ph.D. from the University of Bergen. For the past seven years, he has been working with machine learning and artificial intelligence in Helse Vest IKT.
Recording
You can download the podcast to your mobile on Apple Podcasts, Spotify or Podbean. Search for "Norwegian Centre for E-health Research".