U.S. Markets closed

# This Button Measures Your Calorie Intake As You Eat To Keep Your Diet On Track

IOPScience/Sun

The ebutton is attached to the shirt to monitor meals.

When watching their weight or having health issues, people will often try to keep their own record of the food they are eating.

Humans are very bad at this.

The solution? Build a computer to do it for us. That computer is the eButton, which counts your calories for you.

"Human memory of past eating is imperfect," study researcher Mingui Sun, of the University of Pittsburgh, said in a press release. "Visually gauging the size of a food based on an imaginary measurement unit is very subjective, and some individuals don't want to track what they consume. We're trying to remove the guess work from the dieting process."

An eating problem

The error of self-reporting food intake is about 20 percent, a study in the Journal of the American Dietetic Association found. Humans are bad at guessing volumes, and honesty can be a problem.

The eButton fastens to a person's shirt and automatically takes pictures when the person is eating. These pictures can be used by nutritionists or the person themselves to get a better idea of the number of calories they are taking in.

Essentially, the eButton matches the image it has taken of the food with a 3D shape (which they know the volume of) from a library.

IOPScience/Sun

The eButton takes a picture of the food, removes the background, and then matches the food with a 3D shape.

It can identify the type of food based on its shape, color, and size from a database of foods. There's some limitation here, since cooking habits and dietary cultures around the world vary so much, but the eButton can identify most common foods.

Logging foods

From these two factors, the computer in the chip can calculate the caloric contents of the meal. If a person doesn't finish their meal, the eButton will subtract the leftover food from the beginning volume.

The researchers measured the true volume of 17 different food samples and compared them with the volume the eButton calculated. The average volume error was 3.69 percent. This is a major improvement.

Retrieving the data is a simple process. Eventually the button would be linked to a phone or computer to analyze the data it collects.

The eButton is still in the prototype phase â€” it has a hard time separating foods that look too much like their container, like white rice on a white plate. Sun hopes it will be ready for the mainstream soon.