Study Reveals Lack of Clinical Validation for Nearly Half of FDA-Authorized AI Medical Devices

A multi-institutional study led by researchers from the University of North Carolina (UNC) School of Medicine, Duke University, and others, has revealed that nearly half of the artificial intelligence (AI) medical devices authorized by the U.S. Food and Drug Administration (FDA) lack publicly available clinical validation data. The research, published in Nature Medicine, raises concerns about the effectiveness and safety of these devices, which are increasingly used in healthcare.

Sammy Chouffani El Fassi, an M.D. candidate at the UNC School of Medicine and research scholar at Duke Heart Center, along with Gail E. Henderson, Ph.D., a professor in the UNC Department of Social Medicine, led the analysis of over 500 AI medical devices approved by the FDA. Their findings indicated that 226 devices, approximately 43%, did not have publicly accessible clinical validation data.

Concerns Over Clinical Effectiveness

"Although AI device manufacturers often highlight FDA authorization as a mark of credibility, clearance doesn’t necessarily mean that these devices have been thoroughly evaluated for clinical effectiveness using real patient data," Chouffani El Fassi stated. This observation underscores the central concern that while these devices have regulatory clearance, their actual performance in clinical settings may not be adequately validated.

Since 2016, the FDA has significantly increased its authorization of AI medical devices, from two to 69 annually, reflecting the rapid integration of AI technologies into healthcare. These devices are primarily designed to assist physicians in diagnosing abnormalities in radiological imaging, analyzing pathological slides, dosing medications, and predicting disease progression.

The Challenge of Validation

AI technologies rely on complex algorithms trained on extensive datasets to perform tasks traditionally requiring human expertise. Ensuring these technologies accurately process and analyze new, unseen data is critical for their reliability and safety. The study points out that the FDA's latest draft guidance, published in September 2023, lacks clarity in distinguishing among different types of clinical validation studies, potentially leading to inconsistencies in the evaluation and approval process.

Of the analyzed devices, 144 underwent retrospective validation, 148 were prospectively validated, and only 22 were validated through randomized controlled trials—the gold standard in clinical research. Some devices even used "phantom images"—computer-generated data rather than real patient data—which do not meet the criteria for clinical validation.

Call for Clearer Standards

In response to these findings, the researchers are advocating for clearer standards in clinical validation. They emphasize the need for the FDA to differentiate among retrospective studies, prospective studies, and randomized controlled trials, given the varying levels of scientific evidence these methods provide. The team has shared their findings with FDA directors overseeing medical device regulation, hoping to inform future regulatory decisions.

"We’ve shared our findings with FDA directors overseeing medical device regulation, and we hope our work will inform their regulatory decisions," Chouffani El Fassi said. "We also aim to inspire researchers and institutions worldwide to conduct more rigorous clinical validation studies to ensure the safety and effectiveness of AI in health care."