Gut bacteria show promise as a non-invasive, low-cost screening tool for colorectal cancer (Medical Xpress)

Colorectal cancer remains one of the deadliest cancers worldwide, partly because colonoscopies—though highly accurate—are costly and invasive, leading many to delay screening. Now, researchers at the University of Geneva have developed a machine learning model that analyzes gut bacteria at the subspecies level, achieving a striking 90% accuracy in detecting colorectal cancer from simple stool samples. That’s nearly on par with colonoscopy and better than any other non-invasive method to date. With clinical trials underway, could a low-cost stool test soon reshape how we screen not just for colorectal cancer, but potentially many other diseases linked to the microbiome?