The increasing need for precise dietary monitoring across various health scenarios has led to innovations in wearable sensing technologies. A smart neckband allows wearers to monitor their dietary intake.

Automatically monitoring food and fluid intake can be useful when managing conditions including diabetes and obesity, or when maximizing fitness. However wearable technologies must be able to distinguish eating and drinking from similar movements, such as speaking and walking.

Professor Chi Hwan Lee and colleagues propose a machine-learning-enabled neckband that can differentiate between body movements, speech, fluid and food intake. In a paper published in PNAS Nexus, the neckband’s sensor module includes a surface electromyography sensor, a three-axis accelerometer and a microphone. Together, these sensors can capture muscle activation patterns in the thyrohyoid muscle of the neck, along with body movements and acoustic signals.

In a study of six volunteers, the onboard machine-learning algorithm correctly determined which movements were eating or drinking with an accuracy rate of about 96% for individual activities and 89% for concurrent activities.

The neckband is made of a stretchable, twistable, breathable, mesh-structured textile loaded with 47 active and passive components that can run on battery power for more than 18 hours between charges.

According to Lee, the neckband could be used to calculate insulin dosages for diabetic patients by identifying meal timings—or to aid athletes and other individuals interested in increasing their overall health and wellness.

The innovation of the neckband is a major leap forward in managing chronic diseases and holds the promise of mitigating related health complications on a global scale.

This groundbreaking development was achieved through an interdisciplinary collaboration of multiple internal and external collaborators:

Rita R. Patel, Department of Speech, Language and Hearing Sciences, Indiana University,

Dong Rip Kim, School of Mechanical Engineering, Hanyang University, Seoul, Republic of Korea.

Young L. Kim, Weldon School of Biomedical Engineering, Purdue University

Hyowon (Hugh) Lee, Weldon School of Biomedical Engineering, Purdue University

Fengqing Zhu, Elmore Family School of Electrical and Computer Engineering, Purdue University