The Press Stories
Researchers at Sheba Medical Center have made a groundbreaking discovery in the field of gastroenterology. They have found that machine-learning analysis of capsule endoscopy (CE) videos can accurately predict the need for biological therapy in Crohn’s disease patients.
Crohn’s disease is a chronic condition that causes inflammation in the digestive system, leading to symptoms such as diarrhea, fatigue, stomach aches, cramps, weight loss, and blood in the stool. Currently, gastroenterologists determine the need for treatment by analyzing stool samples. However, the research conducted at Sheba Medical Center shows that artificial intelligence (AI) algorithms outperform human doctors in this regard.
Led by Prof. Uri Kopylov, Prof. Shomron Ben-Horin, and Amit Bleiweiss, the research team published their findings in the Therapeutic Advances in Gastroenterology journal. Using a deep learning model, they analyzed CE videos from 101 Crohn’s disease patients and achieved an impressive 81% accuracy in predicting the need for treatment.
Capsule endoscopy allows for a thorough analysis of the entire digestive system using a microscopic device equipped with a camera and transmitter. The AI algorithms used in this study were able to pick up crucial details in the videos that might be missed by human doctors due to the large amount of visual information.
Previous trials have shown that the AI algorithm can analyze up to 12,000 images in just two minutes with 86% accuracy. Additionally, the AI analysis outperformed an experienced gastroenterologist in both image and data analysis.
The researchers believe that the adoption of AI in clinical practice can greatly improve patient outcomes and open new possibilities for diagnosis and treatment. They are now focused on further validating the technology and implementing it in hospitals and clinics worldwide.
In collaboration with Intel, the next steps include testing the technology in real-world settings and ensuring its successful integration into existing medical systems. With the potential to revolutionize gastroenterology, this research represents a significant breakthrough in the field.
