Revolutionizing Health Wearables: WSU Researchers Unveil Innovative Algorithm for Enhanced Accuracy

Researchers at Washington State University (WSU) have created an exciting new algorithm for wearable health devices. This algorithm is designed to handle missing data from sensors, which could greatly help users in remote or underserved areas.

Wearable devices have become popular for monitoring health. They can track vital signs and provide essential healthcare insights, especially in rural locations. These devices work using sensors and machine learning, but they can struggle when data is incomplete. Data loss can happen for several reasons, like user error or sensor failure.

Ganapati Bhat, an associate professor in WSU’s School of Electrical Engineering and Computer Science, leads this ground-breaking research. He points out that many algorithms assume all data from sensors will be available, which is often not the case. Missing data can severely affect how well these health algorithms function. In critical situations, this could even lead to overlooking serious events, like falls.

Bhat explains, “We may not need a perfect picture of the missing sensor data, as long as we can keep the predictions accurate enough for health-related tasks.” This innovative outlook is what sets their work apart.

The project was made possible by a National Science Foundation CAREER award, and it involves a collaborative team, including graduate students Taha Belkhouja and Dina Hussein, along with associate professor Jana Doppa.

The team’s method has shown promising results in various applications, such as devices to assist paralyzed individuals. Even when several sensors were missing, the accuracy of their algorithm remained intact.

Now, the team is eager to partner with the WSU School of Medicine. They want to apply this algorithm in real-world scenarios, focusing on gesture recognition and activity identification. This could further enhance how we understand and improve health monitoring with wearable technology.