MONDAY, Oct. 20, 2025 — A new AI-driven bite counter is in development to help counter childhood obesity – potentially even tracking kids while they eat and urging them to slow down.

The faster a child takes bites during a meal or snack, the greater their risk for developing obesity, researchers say.

But studying different ways to help kids slow down at mealtime is time-consuming, because researchers must review videos and make note of each bite taken.

To make larger studies possible, researchers developed an artificial intelligence (AI) called ByteTrack that is learning how to count children’s bites during meals.

The system is about 70% as accurate as human counters, according to a report published recently in the journal Frontiers in Nutrition.

As it becomes more accurate, the AI could one day help researchers, parents and doctors identify when children need to slow down their eating.

“When we eat quickly, we don’t give our digestive track time to sense the calories,” researcher Kathleen Keller, chair of nutritional sciences at Pennsylvania State University, said in a news release.

“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,” she said. “Later, you may feel like you have overeaten, but when this behavior repeats, faster eaters are at greater risk for developing obesity.”

Bite rate has become a go-to measure of children’s eating behaviors, said senior researcher Alaina Pearce, a research data management librarian at Penn State.

“Bite rate is often the target behavior for interventions aimed at slowing eating rate,” Pearce said in a news release. “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."

Researchers trained the AI on 1,440 minutes of videos from a government-funded study of overeating in children. The footage included 94 7- to 9-year-olds consuming four meals each on separate occasions.

The team counted bites for 242 videos, and used that information to train the AI model. They then tested the AI on 51 other videos.

The system was 97% successful as a human at identifying a child’s face, but only 70% as successful in identifying every bite.

“The system was less accurate when a child’s face was not in full view of the camera or when a child chewed on their spoon or played with their food, as often happens toward the end of a meal,” lead researcher Yashaswini Bhat, a doctoral student in nutritional sciences at Penn State, said in a news release.

“As one might imagine, this type of behavior is much more common among children than it is with adults,” Bhat said. “Chewing on a utensil sometimes appeared to be a bite, and this complicated the task for the AI model.”

The goal is to develop the AI to the point where it can accurately identify bites in real time, Bhat said.

“One day, we might be able to offer a smartphone app that warns children when they need to slow their eating so they can develop healthy habits that last a lifetime,” Bhat said.

More information

The U.S. Centers for Disease Control and Prevention has more on ways families can prevent childhood obesity.

SOURCE: Penn State, news release, Oct. 16, 2025