Pioneering the next leap in AI-powered wearable health technology

07 Mar 2024 --- A team of US-based researchers are pioneering a device that could track eating behavior like a pedometer tracks steps, highlighting that such technology could offer invaluable insights into dietary habits, potentially revolutionizing nutrition and health approaches.

Funded by a four-year, US$2.4 million grant from the National Institute of Diabetes and Digestive and Kidney Diseases, the project aims to bridge the gap in understanding real-world eating behaviors using AI-enabled wearable technology.

Spearheaded by scientists from the University of Rhode Island (URI) and the University of Texas at Austin, the interdisciplinary initiative promises to provide unprecedented insights into how individuals eat in everyday life.The AI-powered technology could track eating habits like a pedometer tracks steps.

“Eating behavior data collected in labs are most accurate, but people don’t live in labs, so we don’t know what they’re doing in real-world, day-to-day living,” says URI nutrition professor and researcher Kathleen Melanson. “We want to compare the results from our new system to what we already have in the lab to ensure it is collecting data appropriately.”

“The goal is to use this system in research so that we can test our interventions on modifications of ingestive behaviors.”

Wearable nutrition trackers
At the heart of the study are two innovative devices — a conventional smartwatch and a discreet, custom-made sensor positioned on the participant’s jawline. The smartwatch captures arm and wrist movements during eating, while the jawline sensor records chewing motions with remarkable precision, including speed and intensity.

The research unfolds across four progressive phases, each designed to replicate increasingly realistic eating scenarios. Initially, participants will consume prescribed meals under controlled laboratory conditions, allowing researchers to establish baseline data. Subsequent phases will then transition to cafeteria-style and restaurant settings, providing insights into how eating behaviors evolve in less-controlled environments.

In the final phase, participants return to their daily routines while wearing the sensors, enabling researchers to capture real-world eating habits. Key metrics include the speed of eating, chewing patterns, oral processing time and food consumption rates.

“With these kinds of chewing and oral processing behaviors — rapid eating, taking large bites, not pausing between bites, and not chewing sufficiently — people might be over-ingesting calories before the satiety signals have time to develop,” Melanson reveals.

“By assessing these behaviors, we can help develop a system that can be used in interventions to help people adapt their ingestive behaviors to maximize satiety and help with energy intake.”The research looks to increase understanding of eating behaviors and health outcomes.

Understanding obesity
The study will also target individuals at risk of obesity-related health issues, with a focus on Latin American communities in Rhode Island and Texas. These communities face disproportionate rates of obesity and related conditions, making them ideal candidates for research aimed at understanding cultural influences on dietary behaviors.

“The study moves through several successive experiments from an in-lab setting into the real world gradually moving from the internal validity of inferences in a lab-controlled setting to those grounded in the external validity of the real world,” explains psychology professor Theodore Walls, whose research produces statistical tools to make sense of real-time intensive longitudinal health behavior data.

By combining expertise in nutrition, behavioral statistics and engineering, the research team says it seeks to unlock a deeper understanding of eating behaviors and their implications for health outcomes.

“We want to make sure we can make that progression out of the lab in a way that really works in an overall behavioral monitoring approach,” Walls concludes. “Our ‘customers’ are people who want to start clinical trials using this system.”

“This stage is just about the measurement, but later we will move on and test real programs to help people manage their eating behavior.”

Edited by William Bradford Nichols