A medical technology company hopes to pioneer AI-powered ultrasound machines to impart more objectivity to sleep apnea screening. In the future, it may also provide data on whether a therapy will work on a given patient.

How would the field of sleep medicine change if objective airway evaluation could happen in just a few minutes while a patient is awake?

This is a question that one medical device company is hoping to answer with its AmCAD-UO software, cleared by the United States Food and Drug Administration (FDA) in December 2018 to provide data for a physician’s evaluation and monitoring of patients’ airways. The system uses ultrasound technology to scan the neck of patients to look for physical markers of obstructive sleep apnea.

The software then creates an artificial intelligence (AI)-powered assessment report, which shows a prediction of the patient’s risk for severe sleep apnea, or an apnea hypopnea index above 30. The results are calculated based on the scan and other data points, including the patient’s body mass index and the neck circumference.

While there may be a promising future for AmCAD-UO in clinical sleep medicine, there are still questions about how the system will be used.  “For clinical purposes, sleep medicine does not use ultrasound and has no training so that they can be certified in ultrasonography,” says Kingman Strohl, MD, a sleep medicine internist and researcher at the University Hospitals at Cleveland Medical Center, a not-for-profit medical complex in Cleveland, Ohio.

“The application of this would be new and we still need to find out whether it has an appropriate clinical role in decision-making,” says Strohl, who is currently using the AmCAD-UO for research purposes.

Still, developers of the technology have high hopes for the device’s deployment in clinical settings, says James Lee, vice president of business development at AmCad BioMed Corporation.

Medical providers could use the software as a tool to better understand their patients, Lee says, adding that he means not only sleep medicine physicians, but also surgeons—who could employ the system to take a closer look at their patient’s anatomy before an operation. Even dentists can use it to see if an oral appliance is changing the way the upper airways function, says Lee.

Current sleep apnea screeners, such as Stop-BANG and No-SAS, rely on questionnaires in which the patient self-reports their symptoms or physical characteristics, which is not always a reliable method, says Lee. AmCAD-UO incorporates some of the key questions of these written screeners into the OSA risk calculations—but goes a step further by providing a picture of the patient’s physical anatomy, he says. The process takes about 10 minutes per subject from walk-in to result, and the patient is awake during the procedure. According to the FDA, AmCAD-UO must not be used alone for primary diagnostic interpretation.

Multiple images, taken at different angles, generate a 3D reconstruction of the upper airway. This gives physicians and researchers a comprehensive look, detecting airspace and measuring the width within the airway.

“They have post-image processing that allows you to visualize in more than just one plane and you can rotate images, so you can get better 3-D images of the anterior muscles of the throat, the tongue, and the soft palate,” says Strohl.

The imaging is able to clearly indicate where in the patient’s anatomy blockage is occurring during apnea. What’s more, it could potentially evaluate CPAP alternatives to get an idea for their effectiveness, either before or after the patient purchases a therapy device. Ear nose and throat (ENT) physicians can also use the system to evaluate patients for tongue reduction surgery.

The Cleveland Medical Center researchers are currently working with the AmCAD UO to better understand its future applications. “We are still exploring its possibilities,” says Strohl. “Our interest is really in the idea of using ultrasound to assess current and future therapy for obstructive sleep apnea.”

AmCAD UO can easily be deployed in ENT practices, where providers are trained in ultrasonography, says Strohl, but is this system ready for primetime in the clinical sleep medicine subspecialty? “I can’t speculate on what this is, but I think there is a need in sleep medicine to have more procedures, maybe this is one of them,” Strohl says.

Lisa Spear is the associate editor of Sleep Review.