AI Tool Analyzes Capsule Endoscopy Videos to Accurately Predict Outcomes for Crohn’s Disease Patients
By Staff Writers at Hospicedica International
Posted on: July 24, 2023
Crohn’s disease, an inflammatory bowel disease, can cause serious symptoms if not managed properly. There is a clear need for reliable predictors of disease prognosis and response to therapy. Capsule endoscopy, which uses a small device with a camera and transmitter to analyze the entire digestive system, offers a potential solution. However, with approximately 10 to 12,000 images acquired per endoscopic capsule film, the large amount of visual data makes it difficult for physicians to identify all the important details. Now, newly developed artificial intelligence (AI) algorithms can rapidly scan and thoroughly analyze all images produced by the endoscopic capsule within minutes. In addition to identifying inflammation and ulcers, it measures the severity of Crohn’s disease to help doctors choose the best treatment.
Sheba Medical Center (Ramat Gan, Israel) partnered with Intel (Santa Clara, Calif., USA) to develop an AI algorithm-based application that scans and analyzes images produced by endoscopic capsules, detects signs of inflammation and ulcers, and assesses disease severity to inform treatment decisions. With the ability to facilitate early detection of Crohn’s disease, AI platforms are expected to reduce the risk of complications and hospitalization, as well as the need for invasive procedures.
Image: Researchers are using AI to help Crohn’s disease (Photo credit: Sheba Medical Center)
A recent peer-reviewed study validated the accuracy of AI algorithms in predicting treatment outcome for Crohn’s disease and highlighted their potential in guiding biologic therapy decisions. In this study, a team of data researchers and physicians tested this deep learning model using complete capsule endoscopy videos of 101 patients with Crohn’s disease to assess its predictive accuracy for biologic therapy. The AI algorithm showed a staggering 81% accuracy, significantly surpassing analysis by gastroenterologists based on inflammatory index (calprotectin) in stool samples.
This latest research builds on a trial last year that demonstrated the AI algorithm’s ability to process up to 12,000 images in approximately two minutes. In addition, he achieved an accuracy of 86% for imaging and data analysis, compared to his 68% accuracy achieved by experienced gastroenterologists, proving to be a highly effective diagnostic tool. The AI analysis also compared well with an analysis based on the stool inflammatory index (calprotectin).
“Predicting the disease course and patient outcome of Crohn’s disease is one of the most important clinical challenges in treating inflammatory bowel disease. However, this study highlights the potential impact of AI on this process,” said Professor Uli Kopylov, Director of IBD, Department of Gastroenterology, University of Sheva. “By deploying AI in the clinical setting, we can leverage the wealth of knowledge and research in personalized medicine to improve patient outcomes and open the door to new diagnostic and therapeutic possibilities.”
Related Links:
Shiva Medical Center
intel
