Artificial intelligence in radiology – “Risk prediction is very exciting”
“Artificial intelligence is a big topic in my field because images are the perfect target for AI. There is a lot of data, it is fairly structured data, and a large amount of data is composed and collected in an organised way,” said Sophia Zackrisson during a lecture at The Future of Swedish and Danish Life Science event.
She explains that as radiological images are now digital, artificial intelligence is a suitable analytical tool.
To date, around 200 different applications in the field of radiology based on AI have been approved by the US Food and Drug Administration (FDA) or are CE marked. AI techniques are already being used in several hospitals in Sweden, for example, to analyse mammograms.
“We have a shortage of radiologists, and we are needed in so many other areas of diagnostics that we cannot really keep up, and AI will come in handy here.”
Studies indicate that AI, in many cases, provides the same level of accuracy, or in some cases higher, compared to a medical doctor when it comes to detecting cancer. AI is also able to see what radiologists are unable to see with the naked eye.
“However, we cannot let AI work alone,” said Sophia Zackrisson.
At Skåne University Hospital, where Sophia Zackrisson is a physician, the use of artificial intelligence is being explored in areas such as mammography and also in a pilot project for interpreting X-rays of suspected bone fractures.
“We have set up an AI council and a group to work on this to ensure what we call an orderly introduction in healthcare. We define a need and find a validated method that helps us fulfil that need,” says Sophia Zackrisson.
In your opinion, which is the most interesting development in the field?
“I think that risk prediction is very exciting, that AI is not only able to find things that we can already find, but it is also capable of doing things that I, as a radiologist, am unable to do, although the information is contained in the images.”
“Although we still have a long way to go before we reach that point.”
Artikeln är en del av vårt tema om News in English.