Returning from the HIMSS conference, I was protected by gadgets built to guard against human frailties.
Returning from the HIMSS conference, I was protected by gadgets built to guard against human frailties.
Approaching Lambert-St. Louis International airport, a cockpit alarm stood silent watch in the deep background. We had been circling Chicago's Midway airport for more than an hour, when the pilot announced we were running low on fuel and were heading elsewhere. He would have liked to tell us more, he said, but he and the crew were extremely busy, getting us on the ground. If, in the frenetic process of rerouting, ascending, and descending that accompanied our trip to St. Louis, he or the copilot had forgotten to lower the landing gear, an alarm would have sounded.
Hours after that harried diversion to St. Louis, after landing in Chicago and picking my way through Chicago traffic, I slammed on my brakes in an urgent attempt to avoid missing a turn on a Wisconsin highway covered by a foot of still-falling snow. Anti-lock brakes kept me straight.
There are plenty of alarms and safeguards built into modern medical devices: the alarm that sounds when a patient irrationally pulls out a ventilator tube or the interlock that clicks in place to prevent an x-ray overdose to a child. But few manage the interaction of devices.
This is the realm of contextual medicine, where computers, seeded with knowledge to analyze the circumstances surrounding a procedure and information about ongoing events, should protect patients from the "oops" moments that plague us daily. At HIMSS, a demonstrator presented the case history of a bedridden 32-year-old woman on a ventilator. Preparing to take a routine portable chest x-ray, staff at the hospital turned off the ventilator to prevent motion artifact due to respiration, shot the image, then forgot to turn the ventilator back on. She died.
Today's IT developers are wrangling with the formidable problems of capturing and efficiently presenting patient data to staff. But that alone will not be enough to prevent human error. Just as computers onboard airliners know the plane is landing and the landing gear is not in place, or brakes on a car know to pump repeatedly because the car is skidding, healthcare IT systems eventually must be programmed with a contextual understanding of medicine. It's the next step for medicine. It's a big one, but it has to be taken, or the errors dogging modern medicine will not go away.
Four Strategies to Address the Tipping Point in Radiology
January 17th 2025In order to flip the script on the impact of the radiology workforce shortage, radiology groups and practices need to make sound investments in technologies and leverage partnerships to mitigate gaps in coverage and maximize workflow efficiencies.
Can Generative AI Facilitate Simulated Contrast Enhancement for Prostate MRI?
January 14th 2025Deep learning synthesis of contrast-enhanced MRI from non-contrast prostate MRI sequences provided an average multiscale structural similarity index of 70 percent with actual contrast-enhanced prostate MRI in external validation testing from newly published research.
Can MRI-Based AI Enhance Risk Stratification in Prostate Cancer?
January 13th 2025Employing baseline MRI and clinical data, an emerging deep learning model was 32 percent more likely to predict the progression of low-risk prostate cancer (PCa) to clinically significant prostate cancer (csPCa), according to new research.