In a wide-ranging interview, Prof. Alexander Hann of the University Hospital Würzburg shared that while AI has improved adenoma detection rates in colonoscopies, the expected real-world impact—like reduced mortality—remains statistically insignificant. Despite over 40 randomized trials, the evidence has not led to widespread guideline endorsement. Concerns around over-reliance on AI, false positives, and “de-skilling” persist. Hann stresses that for AI to truly transform GI care, interdisciplinary research, robust multi-center datasets, and better endpoints are key.
eHealth: Coding and AI in gastroenterology, where the breakthrough is missing (Heise Online)
0