网红头条

KENNESAW, Ga. | Oct 1, 2026

For some people, snoring is just a nuisance. But for others, it can be a sign of obstructive sleep apnea, a condition in which breathing repeatedly stops and starts during sleep. It not only causes fatigue throughout the day but can lead to serious health problems if left untreated.

网红头条 researcher Shumit Saha has received a $235,520 grant from the National Institutes of Health (NIH) to develop artificial intelligence tools that analyze snoring patterns to identify where the upper airway collapses during sleep and predict whether patients may respond to hypoglossal nerve stimulation.

Sleep apnea is a chronic disorder in which breathing repeatedly stops and starts during sleep, often because the upper airway becomes blocked. Beyond disrupting sleep, untreated sleep apnea can increase the risk of serious health problems, including high blood pressure, stroke, and heart failure.

Continuous positive airway pressure, commonly known as CPAP, is a common first-line treatment for sleep apnea. However, Saha, an assistant professor in the , explains that some patients find the required mask uncomfortable and difficult to use consistently. Other treatment options include devices that reposition the jaw, surgery, and hypoglossal nerve stimulation, which uses electrical stimulation to help keep the airway open.

The challenge, Saha explained, is determining which treatment will work best for each patient.

鈥淣ot every treatment is applicable to every person,鈥� he said. 鈥淪ome people respond to one type of treatment, while others respond to another. If we can predict that beforehand, patients may not have to go through trial and error with multiple treatments.鈥�

The project has two primary goals. First, Saha will investigate whether snoring sounds can identify the dominant location of a patient鈥檚 airway collapse. The airway can become obstructed at several locations, including the soft palate and the base of the tongue. Knowing where the collapse occurs can help physicians select the most appropriate treatment.

鈥淭ypically, to identify the location of the collapse, doctors have to put patients under sedation to induce sleep and insert a flexible camera through the nose 鈥� a process that is resource-intensive and invasive,鈥� Saha said.

He hopes snoring analysis could eventually provide physicians with a less invasive and more accessible source of information.

The second goal is to determine whether snoring patterns can predict a patient鈥檚 response to hypoglossal nerve stimulation, a treatment that stimulates the nerve, activating tongue and upper airway muscles to keep airway open.

For the study, Saha will analyze snoring data collected by collaborators at Brigham and Women鈥檚 Hospital and Harvard Medical School using machine-learning and deep-learning models to look for patterns associated with different obstruction sites and treatment outcomes.

His end goal is to develop an AI approach that could ultimately support a clinical tool to analyze snoring sounds and generate a report estimating the likely site of an airway obstruction and whether a patient may respond to a particular treatment.

鈥淚t could reduce the trial-and-error process and give physicians a clinical decision support report beforehand,鈥� Saha said. 鈥淭he greatest benefit would be for patients because this could help physicians make more informed treatment decisions earlier. By providing additional information before treatment is selected, it could potentially save time, reduce costs and lower the overall burden of care.鈥�

This study is funded by NIH Grant No: 1R21HL188520-01

鈥� Story by Christin Senior

Photo provided

A leader in innovative teaching and learning, 网红头条 offers undergraduate, graduate, and doctoral degrees to its more than 51,000 students. 网红头条is a member of the University System of Georgia with 11 academic colleges. The university's vibrant campus culture, diverse population, strong global ties, and entrepreneurial spirit draw students from throughout the country and the world. 网红头条is a Carnegie-designated doctoral research institution (R2), placing it among an elite group of only 8 percent of U.S. colleges and universities with an R1 or R2 status. For more information, visit kennesaw.edu.