Pavilion Publishing and Media Ltd
Blue Sky Offices Shoreham, 25 Cecil Pashley Way, Shoreham-by-Sea, West Sussex, BN43 5FF, UNITED KINGDOM
Tel: +44 (0)1273 434 943
Email: [email protected]
Scientists have created an artificial intelligence (AI) tool which could help to improve early detection of colorectal cancer.
The tool has been designed to identify flat bowel polyps (small tissue growths that form on the lining of the colon) which are typically very difficult to detect.
Are polyps dangerous?
While most polyps are benign, they contain abnormal cells or cells that may become abnormal over time. For this reason, they are usually removed when found on a colonoscopy to eliminate the chance of the polyp becoming cancerous.
Polyps can be flat, raised or on a stalk. Larger polyps (known as adenomatous polyps or AP) which often resemble a mushroom, are more common and easier to detect.
Comparatively, flat polyps, including flat lesions and sessile serrated lesions (SSLs), are much harder to spot and are also more likely to develop into an aggressive cancer.
The AI tool was able to identify polyps more accurately and more quickly compared to expert colonoscopists
Researchers from the UCL, UCLH and UCL-spinout Odin Vision therefore set out to develop a tool that could be used to identify these flat, harder to spot polyps.
They upgraded the CADDIE (Computer Aided (AI) Device used in Endoscopy) algorithm (which was already able to spot AP) to identify flat lesions and SSLs. They did this by creating four video test datasets that included images of the most challenging flat polyps.
The algorithm was able to identify a 79.5% of the flat polyps, a significantly higher percentage than expert (37%) and trainee (11.5%) colonoscopists. It was also able to do this at a much faster rate.
Across the other datasets, the algorithm was able to detect nearly 100% of polyps.
Colorectal cancer set to become more prevalent
The researchers say the tool could help to reduce colorectal cancer – the second leading cause of cancer deaths in the UK after lung cancer.
Lead author Dr Omer Ahmad (Wellcome / EPSRC Centre for Interventional & Surgical Sciences, UCL) said: “We spent considerable time working with expert colonoscopists to develop the video dataset of extremely challenging polyps that could easily have been missed.
“These polyps were incredibly subtle in appearance and highlight the importance of developing novel technologies to assist colonoscopists during the procedure and improve detection rates.
“We hope that this major system upgrade will support clinicians in the fight against this common cancer, which is set to become more prevalent over the next few years.”