Researchers cite safety concerns after uncovering 'harmful behavior' of fracture-detecting AI model

"Few preclinical artificial intelligence research studies have addressed these concerns; for example, external validation--an assessment of the ability …