Crohn’s disease is an inflammatory bowel condition that, if left unchecked, can spread and cause severe symptoms. However, early detection can not only help alleviate symptoms but, in some cases, result in remission.
While the exact cause of Crohn’s disease is still debated until recently the main suspects were diet and stress. However, recent discoveries suggest these factors only contribute to the development of the condition. As of now, it is believed that Crohn’s is caused by an immune system response to bacteria, although we cannot yet pinpoint the culprit in question. Several studies also point to genetic factors, and it is believed that the disease is more likely to develop in people with a family history of Crohn’s.
As the affected area of the bowel differs from patient to patient, symptoms vary and can include diarrhea, fever, bloody stool, fatigue, mouth sores, and weight loss. Since Crohn’s disease has no cure at the moment, treatment involves reducing inflammation and restricting further complications. Therapies usually involve anti-inflammatory drugs, immunosuppressants, antibiotics, special diets, and in some cases surgery. Regardless of the treatment chosen, early diagnosis is key to ensuring the best results.
When it comes to diagnosing Crohn’s disease, one of the best tools in a gastroenterologist’s arsenal is the endoscopic capsule, also called the “camera pill,” which can take 10,000-12,000 images as it passes through the bowel. However, thoroughly reviewing this massive trove of images is challenging and time-consuming. With this issue in mind, Sheba’s ARC (Accelerate, Redesign, Collaborate) Innovation Center, which unites leading medical and academic institutions worldwide with innovative startups and strategic industry partners to develop, pilot, and roll out game-changing healthcare solutions, has partnered with Intel to develop an application that can quickly scan the images using an artificial intelligence algorithm. The new algorithm can go through all the images produced by the endoscopic capsule and analyze them thoroughly within minutes, spotting inflammations and ulcers, as well as assess the severity of the condition to help decide on the optimal treatments.
“ARC focuses on advanced technologies with the goal of impacting medicine in Israel and globally through digital health solutions,” explains Dr. Eyal Klang, Head of the Sami Sagol AI (Artificial Intelligence) Hub at ARC Innovation Center. “The collaboration with Intel exceeded our expectations and is further proof of the contribution of artificial intelligence to the medical field. Over the next decade, we will see more and more advanced algorithms entering hospitals and supporting doctors’ work.”
According to a joint study, the algorithm achieved better accuracy compared to the traditional, manual approach – averaging 86% accuracy within a short, two-minute analysis compared to 70% accuracy, while the most accurate stool biomarker had 74% accuracy.
As early detection is the key to battling Crohn’s disease, the new platform is expected to decrease the risk of medical complications and hospitalization, as well as the number of invasive procedures. It was already presented at the Israeli Association of Gastroenterology Conference, as well as the annual meeting of the European Inflammatory Bowel Diseases Organization, and is slated for full deployment at Sheba in the next few months. In the future, the platform is expected to expand to other Israeli medical centers, and then, hopefully, to healthcare institutions across the globe.
