Crohn’s disease is a long-term condition that causes inflammation in the digestive system, often leading to pain, diarrhea, fatigue, and weight loss.
One of the ways doctors check how serious the disease is involves a procedure called endoscopy. This allows them to look inside the intestines using a small camera.
But judging how bad the inflammation is can be tricky. Doctors may have different opinions based on what they see, and the current tools to measure it are not always accurate or consistent.
A new study suggests that artificial intelligence, or AI, might help solve this problem. Researchers from the University of Michigan have shown that a computer program can look at endoscopy images and detect signs of Crohn’s disease just as well as—or even better than—human doctors.
The study was published in a medical journal called Clinical Gastroenterology and Hepatology.
In the study, two doctors looked at nearly 4,500 images taken from endoscopy videos. They marked where they saw ulcers, which are a sign of inflammation. These same images were also examined by an AI system trained to detect ulcers. The goal was to see how closely the AI’s findings matched the doctors’.
Interestingly, the AI did a better job matching the “ground truth”—the actual presence of ulcers—than the two doctors did when compared to each other. This suggests that AI could help make endoscopy results more reliable.
The AI system also produced results that closely matched a common scoring tool used by doctors, called the Simple Endoscopic Score for Crohn’s Disease, or SES-CD. This score helps doctors rate the severity of the disease by looking at factors like how many ulcers there are and how large they are.
But even though SES-CD is widely used, it has its flaws. Different doctors may give different scores for the same patient. Plus, the score doesn’t always capture all the small details of the ulcers or other features of the disease. The researchers believe AI could offer more exact, detailed measurements of what’s happening inside the intestines.
Dr. Ryan Stidham, one of the lead authors of the study, explained why this matters. He said that while experienced doctors often have a “gut feeling” about whether a patient’s condition is getting better or worse, it’s hard to back that up with hard numbers.
With AI, they can have stronger, more objective evidence to support their decisions—especially when choosing costly or risky treatments.
Another big benefit is that AI could help in places where there aren’t many experts in Crohn’s disease. In those areas, AI might guide treatment by offering expert-level analysis of endoscopy videos.
For doctors who are already specialists, it could help them better understand and explain the choices they make. It might also help train new doctors and support the development of new treatments.
Dr. Stidham hopes that this research sparks more interest in finding better ways to track Crohn’s disease. He believes that this AI tool is just the beginning, and that in the future, doctors and AI can work side by side to give patients the best care possible.
This study shows that AI can do more than just assist—it can actually improve the way we diagnose and monitor chronic diseases like Crohn’s. By making evaluations more consistent and detailed, AI has the potential to make life better for many people living with this difficult condition.
If you care about gut health, please read studies about how high blood pressure medicine might affect your gut and origin of Parkinson’s disease is in the gut.
For more about gut health, please read studies that your gut can help lower your blood pressure: here’s how and from shock to awe: a zap in the gut could be the new insulin for people with diabetes.
Copyright © 2025 Knowridge Science Report. All rights reserved.
