Researchers at the University of Cambridge have developed an artificial intelligence tool that could significantly speed up the diagnosis of coeliac disease, an autoimmune condition affecting nearly 700,000 people in the UK. The algorithm, trained on over 4,000 biopsy images from five hospitals, was found to be as accurate as a pathologist in detecting the disease but much faster.
Coeliac disease is triggered by consuming gluten, found in wheat, rye, and barley, and symptoms include stomach cramps, diarrhoea, skin rashes, weight loss, fatigue, and anaemia. Untreated, it can lead to serious complications such as malnutrition, osteoporosis, and an increased risk of certain cancers. Currently, diagnosis involves a blood test followed by a duodenal biopsy, which pathologists examine for damage to the villi in the small intestine.
The AI model can diagnose coeliac disease in under a minute, compared to the five to ten minutes a pathologist typically takes per biopsy. Dr Florian Jaeckle, a co-author of the study, noted that duodenal biopsies are often prioritised lower than cancer cases, leading to weeks or months of waiting. With AI, results could be instant once a biopsy is scanned, eliminating waiting lists.
Professor Elizabeth Soilleux, senior author, said: 'It can take many years to receive an accurate diagnosis, and at a time of intense pressures on healthcare systems, these delays are likely to continue. AI has the potential to speed up this process, allowing patients to receive a diagnosis faster, while at the same time taking pressure off NHS waiting lists.'
The study, published in the New England Journal of Medicine AI, was funded by Coeliac UK, Innovate UK, and others. Dr Bernie Croal, president of the Royal College of Pathologists, called the tool 'very exciting' but stressed the need for investment in digital pathology and training before AI can be safely used in the NHS.
