The inside of a human body can be a disorienting and dynamic landscape to the untrained eye. Doctors spend years training to visually identify subtle signs of disease and have relied on endoscopy – medical imaging via a long tube – to look closely at those hard to reach places. But now there’s a digital revolution going on in the field of gastroenterology and endoscopy that has the potential to greatly increase positive patient outcomes.
“No gastrointestinal (GI) investigation and diagnostic procedure can be complete without endoscopy. And artificial intelligence and machine learning are fundamentally changing it,” says Dr Prateek Sharma, professor of medicine at the University of Kansas School of Medicine. Dr Sharma did his medical schooling in Baroda, and then moved to the US for advanced training in gastroenterology.
AI and ML, Dr Sharma says, are being used to develop software that can automatically detect and flag abnormalities in the GI tract. “The machine is first fed with thousands of images of what colon cancer or stomach cancer looks like, and it learns over time how to accurately detect it. So, when you’re performing an endoscopic procedure and a cancerous or pre-cancerous area comes up, the computer recognises it and alerts you,” he says.
The ability of this technology to identify pre-cancerous growth, which is harder to detect, is one of the biggest advantages. “Late-stage cancer is very difficult to treat. Being able to detect cancer early gives physicians the chance to cure the disease,” Dr Sharma says.
The software is able to do this thanks to the help of computer vision. Computer vision is a field of AI that enables computers to understand the visual world, and promises not just to revolutionise the medical field but also has the potential to usher in everything from self-driving cars to autonomous factories.
The software developed right now can only detect colonic diseases, but Dr Sharma says in the future, the software can be expanded to include different parts of the body such as the liver, stomach, and pancreas. “The use of AI is much more advanced in ophthalmology, radiology and dermatology. The tech is just blossoming in gastroenterology. The first AI device for the field of gastroenterology in the US was approved in April. Currently, I’m involved in at least three trials evaluating AI devices in gastroenterology. The hope is that in the next few years there will be hundreds of devices approved,” says Dr Sharma, who was recently named as the chair of the US AI in Gastroenterology and Endoscopy Task Force. The first-of-its-kind taskforce was appointed by GI Society, which works to improve the lives of people with gastrointestinal and liver conditions.
Dr Sharma will be conducting a virtual masterclass in mid-June on the subject of AI in the domain diagnosis of GI disorders. You can sign up for it by searching for ASGE Masterclass Artificial Intelligence. “The next generation of gastroenterologists can be trained on techniques of both diagnostic and therapeutic endoscopy by using VR and AR images of the GI tract,” he says.