Grant: Project Grant

Swallowing problems are common in hospitalized patients, especially after stroke and in older adults. When swallowing fails, food or liquid can enter the lungs. This is called aspiration and can lead to pneumonia, longer hospital stays, restricted diets, and even death. Many patients silently aspirate without coughing or visible warning signs, making detection difficult. Hospitals use bedside screening tests to identify patients at risk. These tests rely on outward symptoms rather than directly measuring swallowing function. To avoid missing dangerous cases, they are designed to be highly sensitive. As a result, many patients who are actually swallowing safely are referred for unnecessary X-ray tests and placed on restrictive diets, leading to avoidable procedures, higher healthcare costs, and added burden on patients and clinicians. We have developed a handheld device called the Aspirometer to improve swallowing screening. A small sensor placed on the neck records vibrations and sounds during swallowing, and machine learning analysis identifies patterns linked to unsafe swallowing. This approach measures swallowing physiology directly rather than relying only on visible signs. In preliminary studies, the Aspirometer maintained high sensitivity for detecting aspiration while significantly reducing false positives compared to current bedside screening. We will now prospectively validate the device in real hospital conditions to determine whether these results hold and whether improved screening performance leads to fewer unnecessary tests, fewer diet restrictions, and better use of clinical resources. If successful, this technology could provide a safe, low-cost way to detect aspiration at the bedside while reducing unnecessary X-ray referrals. It has the potential to lower pneumonia risk, decrease avoidable procedures, reduce healthcare costs, and help patients eat and drink more safely by supporting more informed clinical decision-making.