FDA clears 2 AI applications for use in imaging studies

The Food and Drug Administration is ramping up the pace in clearing the use of artificial intelligence in healthcare imaging.

The agency provided 510(k) clearance, through which the FDA approves the further use of a technology that has demonstrated substantial equivalence to a currently marketed device.

In one instance, the FDA granted 510(k) clearance to Ultromics, based in the United Kingdom, for an image analysis system that automates cardiac analysis to aid in the early detection of cardiovascular disease.

Ultromics’ system, called EchoGo Core, applies AI to echocardiography, aiming to automate the analysis and quantification of ultrasound-based heart scans. It’s been in development for nearly a decade—the first trial kicked off in 2011. Ultromics was spun out of the University of Oxford in 2017.

EchoGoCore Ultromics
Auto contouring generated by EchoGo, calculating key cardiac parameters

Traditionally, echocardiography has relied on the expert eye of clinicians, who use their experience to measure the anatomical structures and identifying the disease, a potentially time-consuming and highly variable process. By automating the process and applying AI analysis, EchoGo enables clinicians to interpret echocardiograms efficiently and accurately and help them make care decisions.

EchoGo uses AI to calculate left ventricular ejection fraction, the most frequently used measurement of heart function, left ventricular volumes and, for the first time for an AI application, automated cardiac strain.

"We have more developments planned in 2020, including EchoGo Pro,” says Ross Upton, founder and CEO of Ultromics, who predicts the system “will be the first AI system able to predict cardiac disease from echocardiography. We are also planning to expand into other geographic regions, including Europe and Asia."

In addition, the FDA has approved technology from Hologic, a Marlborough, Mass.-based company that uses an AI-powered algorithm that the company contends can reduce mammography read times without compromising image quality, sensitivity or accuracy.

Hologic’s system uses a three-dimensional imaging approach, called tomosynthesis, which provides better results than two-dimensional imaging approaches, particularly for women with dense breasts, the company contends. Hologic’s new core artificial intelligence platform aims to provide advanced cancer detection, operational efficiency and clinical decision support across the breast cancer care continuum.

Hologic’s 3DQuorum technology uses Genius AI-powered analytics to uniquely reconstruct high-resolution 3D data to produce 6-millimeter “SmartSlices.” These analytics identify clinically relevant regions of interest and preserve important features during reconstruction of the SmartSlices.

The company contends that SmartSlices expedite read time by reducing the number of images for radiologists to review, without compromising image quality, sensitivity or accuracy.

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