At Digestive Disease Week 2022, Sravanthi Parasa, MD, gastroenterologist at Swedish Gastroenterology, talks about how artificial intelligence (AI) is used in gastroenterology and what challenges it may pose for clinicians.
It is incredibly important for gastroenterologists to understand the development of artificial intelligence (AI) algorithms and whether the questions they are designed to answer are relevant to their patient population, Sravanthi Parasa, MD, gastroenterologist at Swedish Gastroenterology in Seattle, Washington, said at Digestive Disease Week 2022.
What are some ways that AI is being used to help process medical data?
There are different ways as to how AI is being used. When we think about AI, it's just not images or dictation or the notes that we do, it's a bunch of different things. Simple example, for at least in gastroenterology, where it has already made strides is in computer vision, where we are already have 2 FDA-approved algorithms for polyp detection. That's a vision-based algorithm, but there are several other natural language processing-based documentation software, which can help us read through your pathology reports. A lot of times, we know there's a lot of resources that are needed to report some of our quality metrics, almost like 1 FTE [full-time equivalent] a year, so that's another area that has almost reached maturity.
The third one is recruiting patients into clinical trials using computer vision as a source to screen patients who might be eligible for IBD trials. The other aspect of it is the big data, your electronic health records, how do you make better risk prediction models. The list goes on and on, so there's so many different ways that we can use AI in medicine at this point.
How does AI help clinicians, and what are some challenges clinicians face when using it?
The whole point about AI is it's going to be more augmented intelligence. At this point, it's just going to help us make decisions better. Among people who are already at the 95 percentile, probably you won't see much of a difference because they're already excellent, but it helps with standardization of the quality of care that we can provide.
When we start using AI, for example a commercially-available device, we need to understand how AI will interact with humans. Sometimes we get distracted because of the bounding boxes or something, so it's almost like retraining ourselves to a different environment. It's like if you're used to driving a Ford from 1994 to a Tesla in 2020 or 2022: it's a different system but it's not really hard to adapt to it. I think that's the step up that we need to do. But, more importantly, understanding how these algorithms are made and if that question that the AI is trying to answer is relevant to your patient population becomes very important. That's where we want to know how you can interpret the results and stuff, so that's where a little bit of clinical education in terms of basics of AI becomes important.