At the ECCO 2025 Congress, researchers presented a study showing that machine learning can identify gut microbial biomarkers that differentiate IBD patients—particularly those with Crohn’s disease—from healthy individuals. The study analyzed stool samples from over 3,700 participants, finding that supervised machine learning had the highest diagnostic accuracy (AUC 0.971).
Trending
- Long-term outcomes of top-down therapy versus a conventional step-up strategy in adults newly diagnosed with Crohn’s disease: 5-year follow-up of the PROFILE trial (The Lancet Gastroenterology & Hepatology)
- Low AST and ALT Levels Are Associated with Higher Likelihood of Early-Onset Colorectal Cancer in Adults Under 45 (Epic Research)
- Helicobacter pylori exposure linked to one in five cases of bowel cancer (News Medical)
- FDA Clears CU Anschutz Clinical Trial Using Engineered Immune Cells to Fight Colorectal Cancer (PR Newswire)
- A robot pill the size of a large vitamin swims through a stomach on 6 water jets while a doctor steers it with a game controller, and after $21 million it still cannot be used on a patient (The Pulse)
- GLP-1 users can have ‘terrific results’ with bariatric surgery to lose additional weight (Healio)
- GI faces another round of Medicare cuts under proposed 2027 rules (GI & Hepatology News)
- Fecal Occult Blood Screening Outcomes Adjusted for Contamination Bias and Nonadherence (JAMA Network)
