AI smart glasses reduce operating room medication errors by 99.6%

WASHINGTON, UNITED STATES — AI-powered smart glasses developed by UW Medicine are poised to reduce medication errors dramatically in operating rooms, achieving an impressive 99.6% accuracy rate in detecting drug-swapping mistakes, according to a 2024 study. 

The technology could save countless lives in hospitals worldwide, where medication errors remain a persistent threat to patient safety. Medication errors injure at least 1.3 million Americans each year, according to the World Health Organization (WHO), and similar risks are seen globally. 

The new AI-enabled wearable scans drug labels in real time and instantly alerts clinicians to potential mistakes, offering a critical second set of eyes during high-pressure surgeries.

AI wearables revolutionize medication safety

The study from UW Medicine reveals that smart glasses equipped with AI can nearly eliminate vial-swap errors, which cause 20% of all medication errors. Her AI tool, trained on real OR footage, avoids alarm fatigue by only flagging genuine risks.

The device, designed by assistant professor of anesthesiology and pain medicine Dr. Kelly Michaelsen, scans drug labels in real time and alerts staff to mismatches—acting as a second set of eyes during emergencies.

“Ninety-nine percent of the medications we use are these same 10-20 drugs, and so my idea was that we could train an AI to recognize them and act as a second set of eyes,” Michaelsen says.

Despite existing safeguards like barcode scanners, 90% of anesthesiologists admit to medication errors in their careers, per studies cited by Michaelsen.

Balancing AI’s promise with risks

While experts like Dan Cole, University of California, Los Angeles (UCLA) Health and president of the Anesthesia Patient Safety Foundation, compare the technology’s potential to self-driving cars for surgery.

However, Melissa Sheldrick, a patient advocate whose son died from a medication error, stresses “Technology is an important layer in safety, but it’s just one layer and cannot be relied upon as a fail-safe,” emphasizing that AI is not a sole solution and that there are systemic issues like poor communication between hospital departments that cause these medical mistakes. 

Privacy concerns also loom as AI expands into pediatric care, where dosing errors are rampant. While mistakes don’t occur with every patient, the high volume of daily treatments increases the risk of harmful dosing errors, which can have severe consequences for vulnerable young patients. 

Nicholas Cordella, assistant professor of medicine at Boston University’s Chobanian & Avedisian School of Medicine, warns of over-reliance.

Yet proponents argue passive monitoring reduces cognitive strain in chaotic settings where checklists fail.

The road ahead AI integration in surgeries

Although challenges such as privacy concerns and clinician adoption persist, the team believes that AI could soon become standard, allowing medical staff to concentrate on patients instead of preventable mistakes.

“I think that’s where this kind of wearable technology can really come into play, helping us shave off vital seconds and create more time where we can really focus on the patient,” said John Wiederspan, a nurse anesthetist at UW Medicine.