search.noResults

search.searching

dataCollection.invalidEmail
note.createNoteMessage

search.noResults

search.searching

orderForm.title

orderForm.productCode
orderForm.description
orderForm.quantity
orderForm.itemPrice
orderForm.price
orderForm.totalPrice
orderForm.deliveryDetails.billingAddress
orderForm.deliveryDetails.deliveryAddress
orderForm.noItems
BACKPAGE


Artificial Intelligence Could Transform Colonoscopy Improved ADR, quality reporting and reimbursement likely to follow BY ROBERT KURTZ


A


few years ago, colonoscopes with three camera lenses entered


the market. Until then, gastroenterolo- gists who performed colonoscopies had relied for decades on a single-lens plat- form that provided a 180-degree field of view. The three-lens platform nearly doubled that figure and transformed colonoscopy, says gastroenterologist David Robbins, MD, medical direc- tor of Manhattan Endoscopy in New York, New York.


Artificial intelligence (AI), Robbins says, may have a similar transforma- tive effect on colonoscopy. “It will be only a matter of a few years until we see AI in commercial use, either built into existing hardware or as an adjunct tech- nology. Endoscopy is so image-intense and much more about pattern recogni- tion than decision-making, making it perfectly suited to AI.” Helping lead the charge in bringing AI into operating and procedure rooms for colonoscopy is William Karnes, MD, gastroenterologist. He is work- ing with a team of researchers at the University of California, Irvine (UCI) Health in Orange, California, and AI specialists at Docbot, a company that Karnes co-founded. “The concept is using AI to take care


of gastroenterologists’ pain points and help gastroenterologists perform bet- ter,” Karnes says. “The pain points are documentation, completing the opera- tive report and collecting all quality measures and sending them off to the Centers for Medicare & Medicaid Ser- vices (CMS).”


The quality component, Karnes says, includes how well procedures are performed, the adenoma detection rate (ADR), the quality of the prepara- tion, how often the gastroenterologists reaches the cecum and withdrawal time.


26 ASC FOCUS MARCH 2019| ascfocus.org


it could also find polyps or point out possible polyps you might otherwise have missed? Could that improve your ADR and, therefore, your quality mea- sures and reimbursement from CMS? And what if AI could tell you with high accuracy the pathology of the polyp? We are working on developing highly accurate algorithms for all these areas.” For some gastroenterologists, Rob-


The concept is using AI to take care of gastroenterologists’ pain points and help gastroenterologists perform better.”


—William Karnes, MD


These also are quality measures physi- cians are required to report, he notes. “We envisioned AI watching the


procedure and recording these quality measures in real time,” Karnes says. “Then AI could essentially tell you your withdrawal time; Boston bowel preparation score; and how many pol- yps you found, their size and what tool you used to remove them, which goes into your coding for billing.” Karnes and his associates have even bigger visions for AI. “What if


bins says, the emergence of AI strikes right at the center of their “fear button” because of concern that AI might even- tually replace the need for their ser- vices. He has a different perspective. “There are only about 14,000 gas- troenterologists in this country and we are diagnosing about 140,000 colorec- tal cancers annually,” Robbins says. “The reality is that there are not enough of us to go around. Clear applications for machine learning include helping gastroenterologists identify polyps dur- ing an exam as well as helping physi- cian extenders and possibly non-medi- cally trained technicians in the greater war on cancer. AI has a tremendous role to play in education and practical uses.” Karnes says it will not be long until


AI is available to colonoscopists. “Since our focuses include documentation and reporting to CMS, those do not require US Food and Drug Administration (FDA) clearance. We can go to market as soon as we are happy with the way this product performs, but the moment we want to make a claim that the use of algorithms with polyp detection actu- ally improves ADR, that will require FDA clearance. We will also need FDA clearance for optical pathology.” Such applications are likely a mat- ter of when, not if, Robbins says. “AI is here. We are seeing the value of it in other aspects of our lives, and I think we are ready for it for colonoscopy.”


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30