On July 23, 2024, researchers at the National Institutes of Health (NIH) published a study highlighting the risks and benefits of integrating AI into medical decision-making. The study found that an AI model accurately answered medical quiz questions designed to test health professionals’ diagnostic abilities using clinical images and brief text summaries. However, the AI struggled to describe images and explain its decision-making process, even when it arrived at the correct diagnosis. For instance, it failed to recognize that two lesions on a patient’s arm, presented at different angles, were caused by the same condition. The report underscores the need for further evaluation of AI technology before its clinical implementation. Read the study.
Last Updated on August 8, 2024 by Aimed Alliance
