New algorithms speed time to review collapsed lungs

Register now

The Food and Drug Administration has granted 510(k) clearance for radiology technology that serves as a collection of artificial intelligence algorithms embedded on a mobile X-ray device.

Built in collaboration with UC San Francisco and using a platform from GE Healthcare, the AI algorithms help reduce turnaround times for radiologists in reviewing a suspected pneumothorax, which is a type of collapsed lung.

The goal is to get reviews much quicker to radiologists because presently, a prioritized “STAT” X-ray can be sitting for as long as eight hours for a review. But when a patient is scanned on a device with the package of algorithms, called Critical Care Suite, the system automatically analyzes the images by simultaneously searching for a pneumothorax.

If a pneumothorax is suspected an alert and the original chest X-ray is send directly to the radiologist’s picture archiving and communications systems, known as PACS. “X-ray—the world’s oldest form of medical imaging, just got a whole lot smarter,” says Kieran Murphy, President and CEO at GE Healthcare.

“Clinicians are always looking for clinically proven methods to increase outcomes and improve the patient experience,” adds Rachael Callcut, MD, an associate professor of surgery at UCSF, who helped develop Critical Care Suite. “When a patient X-ray is taken, the minutes and hours it takes to process and interpret the image can impact the outcome in either direction,” she adds. “AI gives an opportunity to speed up diagnosis and change how we care for patients, which could save lives and improve outcomes.”

For critical findings, the algorithms ensure AI results are generated within seconds of image acquisition without depending on connectivity or transfer speeds to produce AI results. The results then are sent to the radiologist at the same time that the device sends the original image, ensuring no additional processing delay.

Jie Xue, President and CEO at the X-ray division of GE, says the embedded AI algorithms offer hospitals the opportunity to try AI without making investments in additional IT infrastructure, security assessments or cybersecurity precautions for routing offsite.

For reprint and licensing requests for this article, click here.