Reuters: Medical device manufacturers are actively integrating AI into their devices. Sometimes it leads to serious injuries

Medical device manufacturers have been actively integrating artificial intelligence into their devices in recent years. They are increasingly being used not only for diagnostics, but also in operating rooms — but this is not always without problems, and sometimes leads to serious injuries.

This is stated in the Reuters material.

In 2021, Acclarent, a division of Johnson & Johnson, added an AI algorithm to its TruDi navigation system, which is used during sinus surgeries. The technology was intended to help surgeons navigate more accurately during procedures.

At that time, the device had been on the market for three years, and the US FDA had received only seven reports of its malfunction and one of patient injury. But after the integration of AI, the number of such reports increased to 100.

Moreover, at least 10 patients suffered serious injuries — including cerebrospinal fluid leaks, skull base injuries, and strokes due to accidental arterial injuries.

Reuters, citing an FDA report, writes that most of the cases are related to errors when TruDi misinformed surgeons about the location of their instruments during operations.

At the same time, journalists point out that the FDAʼs reports on the devices may be incomplete and are not designed to determine the causes of medical accidents, so it is unclear what role artificial intelligence may have played. Two stroke victims have filed a lawsuit alleging that the TruDi systemʼs artificial intelligence contributed to their injuries.

Integra LifeSciences, which acquired Acclarent and the TruDi navigation system in 2024, says the reports “contain nothing other than that the TruDi system was used in the operating room where the adverse event occurred”.

The company also insists that “there is no credible evidence to suggest any causal relationship between the TruDi navigation system, the artificial intelligence technology, and any alleged injuries”.

To date, the FDA has authorized at least 1 350 AI devices, nearly double the number in 2022. And the problems weren’t limited to TruDi. There have been cases where heart monitors have misdiagnosed arrhythmias and ultrasound devices have misidentified fetal body parts.

Researchers from Johns Hopkins, Georgetown, and Yale universities found that 43% of AI device recalls occurred less than a year after approval, which is twice as often as recalls of all FDA-approved devices.

Current and former FDA scientists complain that after massive cuts, the agency faces a shortage of resources and personnel to oversee the wildly growing market for AI devices.

Another problem is that the FDA requires clinical trials for new drugs, but medical devices undergo a different review. Under FDA rules, most AI devices that come to market are not subject to testing in patients. Instead, manufacturers comply with the FDA’s requirements by citing previously approved devices that did not have AI functionality.

Positioning new devices as upgrades to existing ones is a long-established practice, but AI adds uncertainty to that practice. Experts we spoke to believe that the FDA’s current regulations for medical devices are ill-suited to overseeing the safety and effectiveness of AI technologies.

For more news and in-depth stories from Ukraine, please follow us on X.