Diagnostics in radiology get a boost from artificial intelligence
In the years to come, Norwegian hospitals will be using more decision support tools built on artificial intelligence. Clinical research shows the way for how we can train the tools to be the best possible, as well as to learn from their mistakes and improve over time.
In many cases, diagnostic images do not have enough information. Age and gender can sometimes tell more than the MRI images. In the future, we need AI models that combine radiological and clinical information.
The Department of Radiology and Nuclear Medicine at Oslo University Hospital has established an artificial intelligence lab to support their medical research. How do they use medical data in the best possible way to train artificial intelligence as decision support? What steps did they take to establish the AI lab? What pitfalls did they encounter along the way?
Presentation by Atle Bjørnerud, leader of the CRAI centre at Oslo University Hospital and professor II at the University of Oslo.
Bjørnerud cites Alzheimer's as an example of how artificial intelligence can contribute to more comprehensive medical diagnostics. In this case, it is trained to look at X-ray images of the brain. It becomes yet another tool in the toolbox that healthcare professionals can utilise.
Read more about CRAI (Computational Radiology and Artificial Intelligence).
Recording
You can download the podcast to your mobile on Apple Podcasts, Spotify or Podbean. Search for ‘Norwegian Centre for E-health Research’.