AI for Esophageal Diseases: Is It Really Helping?

During a session at the recent Digestive Disease Week annual meeting, the many research applications of artificial intelligence (AI) that can be used in the diagnosis and management of esophageal diseases were discussed, including which applications of AI may or may not impact endoscopic practice.

In this MedPage Today video, Prateek Sharma, MD, of the University of Kansas School of Medicine, offers an overview of his session.

Following is a transcript of his remarks:

Artificial intelligence appears to be a new and upcoming field in gastroenterology. But it’s been around in medicine for a number of years. And a number of subspecialties like radiology, cardiology, oncology, are way ahead in terms of their use of artificial intelligence in their practice.

As far as gastroenterology is concerned, there are a number of areas where it is having an impact and will have an impact, starting from artificial intelligence in your endoscopy unit, in which the smart endoscopes now can help you find polyps, for example, during colonoscopy. In the clinic, you have these chat bots, as you’re aware of, you can type your question and a smart, intelligent chat bot will answer some of those basic questions. So, it frees up some time for the physician in order to do that.

There are AI-powered applications. You can download an app, for example, for bowel preparation or if you have a patient with inflammatory bowel disease, for their symptoms. There are a whole variety of AI devices, which are out there.

First and foremost, you should identify what’s your need from a slew of applications that you have available. And then, of course, the important thing is the cost, right? So, what is it that you can or cannot afford? You have to look into that.

But I think it starts right from supply chain, right? I mean, either what you’re stocking, in terms of your gowns or your stocking your accessories for endoscopy, things like that, there are automated ways in which if they are running out, they can be replenished, so that you don’t have to have a person doing a regular bean counter kind of a thing. So, that’s a very simple application which can be done.

All of this will lead to personalized medicine, which uses big data from the endoscopy health records, from the endoscopy reports, from the electronic records, from patient texts, from my notes, putting it all together, and then using this multimodal data to provide personalized medicine to you.