A new artificial intelligence tool developed by La Trobe University researchers could help identify stage-two bowel cancer patients at risk of relapse, according to a Jul. 28 announcement. The technology, known as SÉMIL (Semantically-Enhanced Multiple Instance Learning), analyzes routine pathology slides using images and written descriptions to determine which patients are more likely to experience recurrent bowel cancer.
The research, published in the journal Gastroenterology, involved analysis of more than 1,600 pathology slides and validation across 1,220 stage-two bowel cancer patients from three independent cohorts and several Australian institutions. Findings indicated that when AI-based assessment was combined with pathologist evaluation, the most accurate risk ratings for stage-two cancer patients were achieved.
Francis Magisson, lead author and PhD candidate at La Trobe University's School of Computing, Engineering and Mathematical Sciences, said the technology assesses tumor growth patterns at the invasive front—a feature important for prognosis but difficult for pathologists to classify consistently. "This information could be used to assist pathologists and clinicians to identify which stage-two cancer patients are at higher risk of relapse and may need closer monitoring or additional treatment such as chemotherapy," Magisson said.
Current Australian clinical guidelines recommend chemotherapy after surgery only for high-risk stage-two bowel cancer patients. More accurate identification of high-risk individuals is considered valuable in determining appropriate treatment plans. Associate Professor Zhen He said SÉMIL could be integrated into existing digital pathology workflows without requiring expensive new tests or tissue samples. "SÉMIL could be integrated into existing digital pathology workflows without requiring expensive new tests or tissue samples to provide more consistent and objective assessments of cancer pathology," He said.
Associate Professor David Williams emphasized that the tool is designed to support clinical decision-making rather than replace it. "Our study shows that AI-based assessment has potential to provide pathologists with an additional layer of information that could help clinicians make more informed decisions about next-stage treatment options," Williams said.
Bowel cancer is Australia's fourth most commonly diagnosed cancer and the second leading cause of cancer death; globally it ranks as the third most common type. The study included collaborators from multiple institutions including WEHI, Monash University, University of Melbourne, UNSW Sydney, Peter MacCallum Cancer Centre and other leading centers.
