New Study Finds Wearables May Reshape Obesity Care
Key Takeaways:
- A new study from Northwestern University demonstrates how wearable devices can identify five distinct overeating patterns in people living with obesity, paving the way for more personalised interventions.
- The HabitSense body camera and NeckSense necklace provide unprecedented yet privacy-conscious insights into real-world eating behaviour.
- Researchers emphasise that overeating is not simply a matter of willpower but is shaped by complex emotional, environmental and behavioural factors.
Rethinking obesity treatment through technology
What if a smartwatch, necklace or discreet camera could sense when someone is about to overeat, and instead gently encourage healthier decisions?
Northwestern University scientists are exploring this idea through a pioneering lifestyle medicine programme that combines wearable technology with behavioural analysis. The approach uses three different devices – a necklace, a wristband and a body-mounted camera – to capture eating habits in natural settings, with privacy firmly safeguarded.
“Overeating is a major contributor to obesity, yet most treatments overlook the unconscious habits that drive it,” explained corresponding author Nabil Alshurafa, Associate Professor of Behavioural Medicine at Northwestern University Feinberg School of Medicine and, by courtesy, of Computer Science and Electrical and Computer Engineering at Northwestern’s McCormick School of Engineering.
Five distinct overeating patterns identified
In the study, published in npj Digital Medicine (part of the Nature Portfolio), 60 adults living with obesity wore the three sensors and logged contextual information – such as mood, activity and social setting – using a smartphone app over a two-week period. The project generated thousands of hours of data, revealing that overeating typically followed one of five recurring patterns:
- Take-out feasting – heavy consumption of delivered or takeaway meals.
- Evening restaurant revelry – social dining leading to excessive intake.
- Evening craving – compulsive late-night snacking.
- Uncontrolled pleasure eating – spontaneous binges driven by enjoyment.
- Stress-driven evening nibbling – grazing triggered by anxiety.
“These patterns reflect the complex dance between environment, emotion and habit,” said Alshurafa. “What’s amazing is now we have a roadmap for personalised interventions.”
A step towards personalised interventions
The findings create a foundation for future clinical practice, in which individuals may be profiled according to their dominant overeating pattern and then matched with tailored interventions.
Lead author Farzad Shahabi, a PhD student in Computer Science and member of Alshurafa’s laboratory, highlighted the significance:
“What struck me most was how overeating isn’t just about willpower. Using passive sensing, we were able to uncover hidden consumption patterns in people’s real-world behaviour that are emotional, behavioural 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.”
HabitSense – A body camera with built-in privacy
The project’s roots date back to when Alshurafa borrowed a police body camera from Northwestern’s campus police. He modified it to record only food-related actions, creating what is now called HabitSense.
HabitSense is the first patented Activity-Oriented Camera (AOC), which uses thermal sensors to activate recording solely when food enters the field of view. Unlike conventional egocentric cameras that capture everything from the wearer’s perspective, AOCs record actions rather than scenes. This innovation preserves bystander privacy while still collecting critical behavioural data.
NeckSense – Recording eating behaviours in real time
Participants also wore NeckSense, a necklace designed by Alshurafa and his team. NeckSense is the first technology able to passively and precisely monitor multiple eating behaviours. It can detect when someone is eating, how many bites they take, their chewing rate and the frequency with which their hand moves to their mouth. This provides researchers with highly detailed insight into real-world eating events.
A wrist-worn activity tracker – similar to a Fitbit or Apple Watch – completed the three-sensor system.
From personal struggles to scientific mission
Alshurafa’s scientific interest in obesity stems from his own personal journey. Throughout his younger life, his weight fluctuated by 40 to 50 pounds, with repeated attempts at dieting often undermined by late-night binge eating in front of the television.
“I tried to turn my personal struggle into a scientific mission that promises to reshape obesity treatment,” he reflected. “By merging computer science, behavioural medicine and a dash of Jane Goodall–style curiosity, we’re working to lead the way toward truly personalised, habit-based health care. This study marks only the beginning of a journey toward smarter and more compassionate interventions for millions grappling with overeating.”
Study team and support
The research team behind this project brought together a wide range of expertise from Northwestern and beyond. Contributors included PhD student in computer science Boyang Wei, HABits Lab research study coordinator Chris Romano, and undergraduate student Rowan McCloskey. They were joined by adjunct faculty members Annie Lin of the University of Minnesota and Mahdi Pedram of the University of North Texas, as well as former Northwestern faculty member Tammy Stump, now at the University of Utah. Jacob Schauer, Assistant Professor of Preventive Medicine, also played a role, alongside computer science PhD student Glenn Fernandes and senior engineer Tanmeet Butani (MS ’23).
The study was funded by the US National Institutes of Health through the National Institute of Diabetes and Digestive and Kidney Diseases.
CCH insight:
This is a fascinating study, which shows how new technologies may be able to provide innovative digital solutions to health issues, in this case identifying behavioural patterns underpinning overeating. These results need to be verified in larger studies, and then interventions trialled to address the different eating patterns, so we are a long way from viable new interventions, but this is an intriguing addition to the development of precision treatments for obesity.
