Wearable sweat monitor uses machine learning to protect workers from dehydration

Machine Learning


Wearable sweat monitor uses machine learning to protect workers from dehydration

The Connection Hydration is worn on the bicep and features microchannels and a sensing module that vibrates when filled with sweat, reminding the wearer to hydrate. Credit: Epicore Biosystems

The first commercially available sweat and hydration monitoring device based on rigorous research from the University of Massachusetts Amherst is poised to help protect “industrial athletes” from dehydration and workplace accidents.

Researchers at the University of Massachusetts Amherst partnered with Epicore Biosystems, a digital health company based in Cambridge, Massachusetts, to test Epicore's new wearable sweat and hydration monitor, Connected Hydration.

Connected Hydration is designed for “industrial athletes” – people whose jobs involve frequent strenuous physical activity in hot and humid environments while wearing personal protective equipment, putting workers at risk for dehydration and workplace injuries.

AJ Aranyosi, chief digital officer and co-founder of Epicore Biosystems, explains that water loss of just 2% of body weight can lead to cognitive decline.

“If you go back through the literature, you’ll find that heat exhaustion and heat stroke are actually the root cause of accidents that are happening in industry,” he says. “Currently, many of the national workplace safety guidelines give general recommendations based on weather conditions and the average person. [but] “There is no such thing as an average human being. The amount of sweat and the composition of sweat vary from person to person.”

That's where Connected Hydration comes in. Connected Hydration empowers workers to proactively address the negative effects of dehydration and heat stroke by providing behavioral insights and a better understanding of their personal sweat and electrolyte loss. The device allows users to monitor hydration levels based on sweat volume and composition. Worn on the biceps, the device contains microchannels and a sensing module that vibrates when the channels fill with sweat, reminding the wearer to hydrate.

“Thirst is very subjective,” Aranyoshi says, “and we know from decades of research with athletes that by the time they feel thirsty, they're already somewhat dehydrated. We also know that even when they feel thirsty, they tend not to drink enough fluids to completely replenish the fluids they've lost. Just feeling thirsty is not a safe way to hydrate.”

The platform connects wirelessly to a mobile phone to provide a detailed view of sweat electrolyte levels over time, allows users to log their fluid intake, and connects to the cloud, allowing employers to view aggregate data from their employees and determine whether they are staying safely hydrated or if intervention is required.

To enable a commercial launch of Connected Hydration, Epicore turned to the core facilities at the University of Massachusetts Amherst’s Institute for Applied Life Sciences (IALS) to obtain the data it needed to refine and validate its product.

“The gold standard for assessing whole-body fluid loss is to measure the difference in dry weight pre- and post-exercise,” says Mike Busa, director of the Center for Human Health Performance at IALS. He explains the level of rigor they used in collecting the data to create a truly closed system. This included urine analysis to test hydration status before the study began, pre-weighed garments, measuring dry weights of people wearing garments of known weights pre- and post-exercise, and tracking fluid intake and urinary fluid loss during the test.

“We can closely control the temperature in the room and monitor the participants using a variety of devices,” says Corina Serviente, operations manager and senior physiologist at the Center for Human Health Performance. “We had the 44 participants wear personal protective equipment, like that worn by industrial workers in the field. We raised the temperature on the indoor calorimeter to 86 degrees.”o “And having participants exercise while wearing multiple devices that measure core temperature, heart rate, fluid loss and activity levels provided the data Epicore needed to develop and test its fluid loss algorithms in a way that field testing couldn't.”

“Some people might lose about a half-pound or a quarter-pound when they sweat,” Aranyoshi adds, “but when you're weighing a 200-pound person and looking for a quarter-pound difference, you have to take into account all the causes of weight change.”

As well as using dry weight to measure sweat loss, the researchers also collected subjects' sweat using a non-electronic device from Epicore called the Discovery Patch. The team also analysed the electrolyte composition of sweat using mass spectrometry.

The data provided to Epicore Biosystems was used to create machine learning algorithms to improve its Connected Hydration technology.

“What we wanted to know was how much sweat was inside the device and how that corresponds to how much sweat is produced throughout the human body,” Aranyoshi said. “We started this study with our initial best guess of how that conversion works, and as a result of this work we were able to derive a new equation that allows us to map the voltage.” [from the device]sweat rate and salinity, ultimately providing more accurate measurements and feedback.”

On the UMass side, the collaboration has been a valuable opportunity for students. “We’ve seen a lot of really cool student efforts come out of this project, and Epicore has been great at supporting those students and allowing them to explore and take ownership of some of the data,” Serviente says.

One of those students is Remi Hirsch, a recent graduate from the School of Natural Sciences' Department of Biology, who used some of the data from this project to write her honors thesis and present it at the American College of Sports Medicine's 2024 Annual Meeting in Boston.

Her study found that in young, healthy adults wearing personal protective equipment, the relative humidity of the environment had no effect on the body's metabolic activity, body temperature, heart rate, or fluid loss during intermittent, moderate-intensity exercise. However, the study showed differences in skin temperature and sweat composition between different humidity conditions, indicating areas for future investigation.

“These results are important in that data from different test sessions can be combined and used to develop and test algorithms,” Busa said. “They also suggest that the body may respond differently to exercise in different environmental relative humidity when wearing personal protective equipment. PPE appears to be non-breathable and creates its own microenvironment, which remains to be tested.”

Courtesy of University of Massachusetts Amherst

Quote: Wearable Sweat Monitor Uses Machine Learning to Protect Workers from Dehydration (June 24, 2024) Retrieved June 24, 2024 from https://medicalxpress.com/news/2024-06-wearable-machine-workers-dehydration.html

This document is subject to copyright. It may not be reproduced without written permission, except for fair dealing for the purposes of personal study or research. The content is provided for informational purposes only.





Source link

Leave a Reply

Your email address will not be published. Required fields are marked *