Algorithm Analyzing Complete Capsule Endoscopy (CE) Videos at Initial Diagnosis Achieved 81% Accuracy, Significantly Higher than Analysis by a Gastroenterologist Using Doctoral Analysis of The Inflammatory Index in Stool Samples
Sheba Medical Center, Israel’s largest medical center, published peer-reviewed research confirming the accuracy of a newly developed AI algorithm for Crohn’s Disease (CD) therapy prediction. The study, published in the Therapeutic Advances in Gastroenterology journal, demonstrates that machine learning analysis of complete capsule endoscopy (CE) videos at initial diagnosis can accurately predict the need for biological therapy.
Led by Prof. Uri Kopylov, Director of the IBD unit at Sheba Medical Center, Prof. Shomron Ben-Horin, Director of the Department of Gastroenterology at Sheba Medical Center, and Intel Engineering Director Amit Bleiweiss, data researchers and Sheba physicians trialed a newly developed deep learning model on complete CE videos from 101 Crohn’s Disease patients to see whether the program could predict the need for biological therapy.
Results from the trial demonstrated that the algorithm was accurate in predicting the need for biological therapy for Crohn’s Disease patients, achieving 81% accuracy. This was significantly higher than analysis by a gastroenterologist using doctoral analysis of the inflammatory index in stool samples (calprotectin).
“Predicting disease course and patient outcomes for Crohn’s Disease is one of the most critical clinical challenges in inflammatory bowel disease treatment. However, this research highlights the potential impact of AI on this process,” said Prof. Uri Kopylov, Director of IBD in the Department of Gastroenterology at Sheba. “By adopting AI in clinical practice, we can begin to use our wealth of knowledge and research in personalized medicine to drive improved patient outcomes and open the door to new possibilities for diagnosis and treatment.”
In Crohn’s Disease, predictors of disease prognosis and response to treatment are still lacking. Capsule endoscopy allows for the analysis of the entire digestive system using a microscopic device equipped with a transmitter and camera. However, every capsule film produced includes approximately 10 – 12 thousand images for interpretation. Due to the large amount of visual information in each video, it is difficult for a doctor to discern all necessary details. These details can be picked up by AI algorithms instead.
This research followed a trial last year, in which the AI algorithm demonstrated it could scan a film of up to 12,000 images in approximately 2 minutes. Additionally, the research found AI to be a highly effective diagnostic tool, producing 86% accuracy in image and data analysis compared to 68% accuracy achieved with the reliance on analysis by an experienced gastroenterologist. In this research, AI analysis was also compared to doctoral analysis of the inflammatory index in stool (calprotectin).
“The findings from this research are further proof of the powerful impact that AI can have in transforming our health systems and driving positive patient outcomes,” said Dr. Eyal Klang, Head of the Sami Sagol AI (Artificial Intelligence) Hub at the ARC Innovation Center. “Building on our successful collaboration with Intel, we are looking ahead to further validations of this technology and seeing it implemented in hospitals and clinics worldwide.”