Search
By Andrew Barbarino, Sr. Product Manager, Endoscopy, FUJIFILM Healthcare Americas Corporation
Not long after colorectal cancer (CRC) screening recommendations were introduced across the Western world in the mid-1990’s, colonoscopy emerged as a cornerstone of disease prevention and surveillance. Thanks to widespread adoption of colonoscopy, CRC incidence and death rates have steadily declined throughout the 2000’s. One comprehensive meta-analysis showed a 62% relative risk reduction in mortality and another study found that colonoscopy reduced the risk of CRC death by 50%. It’s no wonder colonoscopy is considered the “gold-standard” for CRC screening, but a closer look at the statistics reveals that even a “gold standard” can be buffed up in the era of Artificial Intelligence (AI).
Despite progress, CRC remains a leading cause of cancer-related morbidity and death, accounting for nearly 10% of all cancer cases and deaths globally (Global Cancer Statistics 2022). Although colonoscopy is highly effective for the early detection and removal of polyps before progression to cancer can occur, it’s not a perfect solution.
To help address these challenges in colonoscopy, AI is being deployed to improve key quality indicators, like polyp detection rate, and to aim for a higher standard of quality across providers, settings, and patient demographics. The emergence of AI-powered technology for colon polyp detection support has been described as a potential gamechanger in the fight against CRC, but we’ve only scratched the surface in terms of AI possibilities in the field of endoscopy. Next generation AI in colonoscopy may support diagnosis in addition to detection, while other AI-powered software aims to improve procedural efficiency and reduce administrative burden on providers.
Fujifilm has a proven track record in AI development alongside advancements in medical imaging, and we continue to invest in an AI-powered future. We established and operate twoBrain(s) Creative AI Centers leveraging state-of-the-art supercomputers for AI development through joint research with academic institutions. In 2024, Fujifilm launched CAD EYE®, our novel AI detection system for endoscopic imaging. Studies show that CAD EYE is helping clinicians to decrease missed rates and improve adenoma detection rates. Today, leading physicians like Dr. Daniel von Renteln, associate professor of Medicine at University of Montreal (UdeM) are utilizing CAD EYE to elevate performance and quality in colonoscopy.
At a recent industry event, Dr. von Renteln shared industry insights as well as his team’s clinical experiences with CAD EYE. He began by stating that adenoma detection rate (ADR) targets in screening colonoscopies were recently revised for North America—going from a minimum of 25% to 35%, a significant increase. Now more than ever, it’s critical for clinicians to have the right detection tools—and AI is making a powerful difference. Dr. von Renteln noted that many published randomized controlled trials consistently show the use of Fujifilm’s CAD EYE increases adenoma detection rate—”our most important quality metric for colonoscopy.”
He also discussed the successful use of CAD EYE at his hospital, the University of Montreal Medical Center (CHUM). While the hospital already had a standard ADR of 38%, once they implemented an optimized technique of Fujifilm’s CAD EYE plus LTI, the CHUM team moved up to an almost 50% adenoma detection rate. That’s a tremendous gain—an 11% increase in ADR and about a 30% relative risk reduction for the patients to develop a colorectal cancer once they leave the endoscopy unit.
While no one is encouraged to rush through a colonoscopy, Dr. von Renteln’s in-house studies also found that when a procedure is done a bit faster, CAD EYE serves as a “safety net,” helping to ensure that lesions are not missed.
Moreover, Dr. von Renteln pointed out the easy operation of CAD EYE, which is seamlessly integrated in the Fujifilm endoscope. The CAD EYE detection tool can be activated and deactivated by a simple push of a button on the endoscope or directly on the processor. The new ADR targets can help save lives and CAD EYE can help clinicians meet those targets. In addition to CAD EYE, Dr. Renteln briefly discussed Fujifilm’s SCALE EYE technology, a real-time virtual measurement tool that uses a Class 1 laser and specialized software to estimate the size of lesions during colonoscopy. In his pilot studies, he found that visual estimation of polyps resulted in about 67% relative accuracy while size estimation using SCALE EYE was 85% accurate.
“Implementation and understanding how to use these technologies consistently is key
for good results in our day-to-day practice.”
You can watch Dr. von Renteln’s presentation in the video below.
Andrew Barbarino, Sr. Product Manager, Endoscopy, FUJIFILM Healthcare
Americas Corporation.