May 8 2024Reviewed by Lexie Corner
Chi Hwan Lee and his colleagues developed a wise neckband that tracks consuming and ingesting habits. This machine-learning-powered system makes use of sensors to tell apart between these actions and related actions like speaking and strolling, providing the potential for managing diabetes and enhancing general well being. The examine was printed within the journal PNAS Nexus.
The sensor module, positioned on the thyrohyoid muscle, contains a 45-degree pre-curved design and comfortable, waterproof encapsulation. Image Credit: Park et al.
Wearers of good neckbands can monitor how a lot meals they eat. When maximizing health or managing sicknesses like diabetes and weight problems, routinely monitoring meals and hydration consumption is likely to be useful. However, consuming and ingesting have to be distinguished from associated motions like talking and strolling through wearable expertise.
Chi Hwan Lee and colleagues suggest a neckband with machine studying capabilities that may distinguish between speech, physique language, hydration, and meals consumption. The neckband’s sensor module features a microphone, a three-axis accelerometer, and a floor electromyography sensor.
Combined, these sensors can file auditory knowledge, physique motions, and muscular activation patterns within the neck’s thyrohyoid muscle. With an accuracy charge of virtually 96 % for particular person actions and 89 % for concurrent actions, the machine-learning algorithm precisely recognized which actions in analysis involving six individuals had been associated to consuming or ingesting.
Made of breathable, mesh-structured, versatile, and twistable fabric, the neckband has 47 lively and passive elements that may function on battery energy for over 18 hours with out recharging.
The neckband may also help athletes and different individuals excited by enhancing their basic well being and wellness. It will also be utilized in a closed-loop system with a steady glucose meter and an insulin pump to assist diabetic sufferers decide when to take their insulin dosages.
Journal Reference:
Park, T., et al. (2024) A machine-learning-enabled good neckband for monitoring dietary consumption. PNAS Nexus. doi.org/10.1093/pnasnexus/pgae156.
Source: https://academic.oup.com/pnasnexus
https://www.azosensors.com/news.aspx?newsID=15828
