UNIVERSITY PARK, Pa. — Researchers at Penn State University have unveiled a groundbreaking artificial intelligence (AI) model designed to monitor children’s eating habits, potentially offering a new tool in the fight against childhood obesity. The study, conducted by the Penn State Department of Nutritional Sciences, reveals that children who eat faster are at a higher risk of developing obesity. However, tracking a child’s bite rate has traditionally been a labor-intensive process, often limited to small-scale laboratory settings.
The new AI model, developed in collaboration with the Department of Human Development and Family Studies, aims to make bite rate counting feasible for larger studies and diverse environments. A pilot study, recently published in Frontiers in Nutrition, indicates that the AI system is currently about 70% as effective as human counters. Despite needing further refinement, researchers believe this technology holds promise for identifying when children need to adjust their eating speed.
Understanding the Link Between Eating Speed and Obesity
Eating quickly can prevent the digestive system from signaling fullness in time, leading to overeating, according to Kathleen Keller, professor and co-author of the study. “The faster you eat, the faster it goes through your stomach, and the body cannot release hormones in time to let you know you are full,” Keller explained. This behavior, if repeated, can significantly increase the risk of obesity.
Previous research from Keller’s group has shown that a faster bite rate, particularly when combined with larger bite sizes, correlates with higher obesity rates in children. Additionally, larger bite sizes have been linked to an increased risk of choking.
“Bite rate is often the target behavior for interventions aimed at slowing eating rate,” said Alaina Pearce, research data management librarian and co-author of the study. “This is because bite rate is a stable characteristic of children’s eating style that can be targeted to reduce their eating rate, intake, and ultimately risk for obesity.”
Leveraging AI for Healthier Eating Habits
To tackle the challenge of measuring bite rates more efficiently, Yashaswini Bhat, a doctoral candidate and lead author of the study, spearheaded the development of the AI bite counter. Bhat collaborated with Timothy Brick, an associate professor at Penn State, to create a system capable of identifying children’s faces in videos and detecting individual bites.
The researchers utilized 1,440 minutes of video footage from Keller’s Food and Brain Study, funded by the National Institute of Diabetes and Digestive and Kidney Diseases. This footage included 94 children, aged seven to nine, consuming meals under controlled conditions. By analyzing the videos, the team trained the AI model to recognize bite events, which was then tested against a separate set of videos.
“The system we developed was very successful at identifying the children’s faces,” Bhat noted. “It also did an excellent job identifying bites when it had a clear, unobstructed view of a child’s face.”
Challenges and Future Prospects
Despite its initial success, the AI system, named ByteTrack, still faces challenges. It was approximately 97% as effective as humans in recognizing faces but only 70% as successful in identifying bites. The system struggled when a child’s face was not fully visible or when children engaged in behaviors like chewing on utensils.
The researchers are optimistic about the potential of ByteTrack. With further training, the system could more accurately identify bites and disregard non-eating actions, such as sipping beverages. The ultimate goal is to develop a robust system that can function in real-world settings, potentially leading to a smartphone app that helps children develop lifelong healthy eating habits.
“One day, we might be able to offer a smartphone app that warns children when they need to slow their eating,” Bhat said. “This could help them develop healthy habits that last a lifetime.”
Implications for Future Research and Health Initiatives
This research, funded by several national institutes and Penn State organizations, underscores the importance of continued support for scientific innovation. At Penn State, researchers are tackling real-world problems that affect health and quality of life globally.
However, ongoing federal funding cuts pose a threat to such progress. The advancements made by studies like this one highlight the critical role of federal support in driving innovation that enhances public health, strengthens industries, and bolsters the economy.
As the research community navigates these challenges, the development of AI tools like ByteTrack represents a promising step forward in addressing childhood obesity. With further refinement and support, these technologies could play a crucial role in fostering healthier generations.
