Wearable sensors are already used in a multitude of fitness and health apps, and now a group of scientists are focused on combatting obesity with around-the-clock insights.
Researchers at Northwestern University used wearable sensors to track eating behaviors, with each study participant wearing a necklace, wristband and body camera to capture real-world eating behaviors. A smartphone app was used to track meal-related mood and context, such as who they were with and what they were doing, for two weeks.
With video and sensor data across 60 adults with obesity, the study resulted in thousands of hours of insights. Overeating is not one-size-fits-all, the researchers said, and they did describe five distinct patterns of overeating the data revealed:
- Take-out feasting and gorging on delivery and take-out meals.
- Evening restaurant reveling, including social dinners that lead to excess foot intake.
- Evening craving with late-night snacks.
- Pleasure eating with spontaneous and joyful binges.
- Stress-driven nibbling in the evening, with grazing fueled by anxiety.
“These patterns reflect the complex dance between environment, emotion and habit. Now we have a roadmap for personalized interventions,” said Nabil Alshurafa, associate professor of behavioral medicine and computer engineering at Northwestern.
His team is already working with clinicians to pilot trials of behavior-change programs based on the findings.
“What struck me most was how overeating isn’t just about willpower,” said lead study author Farzad Shahabi, a PhD student in Alshurafa’s lab. “Using passive sensing, we were able to uncover hidden consumption patterns in people’s real-world behavior that are emotional, behavioral and contextual. Seeing the patterns emerge from the data felt like turning on a light in a room we've all been stumbling through for decades. Our long-term vision is to move beyond one-size-fits-all solutions and toward a world in which health technology feels less like a prescription and more like a partnership.”
The bodycam in the study was programmed to only record the wearer’s food-related actions, for bystander privacy. Called HabitSense, it is a patented Activity-Oriented Camera that relies on thermal sensing to trigger recording when food enters the field of view. It records the activity and not the scene.
The necklace was designed by Alshaurafa and his team and called NeckSense. It is designed to precisely and passively record multiple eating behaviors, including when people eat, how fast they chew and how many bites they take as well as how many times their hands move to their mouths. It relies on an accelerometer and gyroscope to detect if someone leans back to drink or leans forward to eat. It also measures activity which increases when a person moves and falls flat when they stop moving.
Study participants also wore a wrist-worn activity tracker similar to a FitBit or Apple Watch.
Alshurafa said he struggles with his own weight and had fluctuated by 40 to 50 pounds most of his young life. He tried different diets and said he got into a binge cycle when eating late at night watching TV.
“I tried to turn my personal struggle into a scientific mission,” he said.
The study is published in the journal npj digital medicine.
It is titled “Unveiling overeating patterns within digital longitudinal data on eating behaviors and contexts.” Other Northwestern authors include Boyang Wei, Chris Romano, Rowan McCloskey, Annie Lin, Mahdi Pedram, Jake Schauer and Tammy Stump. PhD student Glenn Fernandes and senior engineer Tanmeet Butani contributed to the hardware system.
Funding for the study was provided by the National Institutes of Health’s National Institute of Diabetes and Digestive and Kidney Disease (grant 5K25DK113242).
In a video, the study team also simulated takeout feasting with sensors, minus the allure of a sports bar with exciting major league games displayed on multiple monitors: