Health & Wellbeing

NutriRay3D uses laser light and your phone to count calories

NutriRay3D uses laser light and your phone to count calories
The NutriRay3D prototype in action
The NutriRay3D prototype in action
View 3 Images
A rendering of what the final NutriRay3D scanner should look like
1/3
A rendering of what the final NutriRay3D scanner should look like
The NutriRay3D prototype in action
2/3
The NutriRay3D prototype in action
In lab tests of the current NutriRay3D prototype, it estimated nutritional content of various foods with an accuracy rate of between 87.5 and 91 percent
3/3
In lab tests of the current NutriRay3D prototype, it estimated nutritional content of various foods with an accuracy rate of between 87.5 and 91 percent
View gallery - 3 images

There are already plenty of apps that let people estimate how many calories are in the foods they're eating. However, most of these programs require users to either guess at their portion sizes, or actually weigh the food. That's where the University of Washington's NutriRay3D comes in. It's a smartphone device/app combo, that uses lasers to ascertain how many calories are sitting on the plate.

Once it's fully developed, NutriRay3D's hardware component should consist of a small laser-scanning module that's simply plugged into an existing phone. Utilizing that module, users start by projecting a grid of light points onto individual food items. Using the phone's camera, the app notes how those dots align with one another on the surface of each item, and from there is able to map the size and shape of the food.

The app is also able to identify what many basic foods are simply based on their appearance. In cases where it can't, however, users can simply tell it by speaking into the phone's mic. Once the app knows what the food is and how much of it is present, it's able to consult a database of over 9,000 foods to calculate its caloric content – along with other nutritional information.

In lab tests of the current NutriRay3D prototype, it estimated nutritional content of various foods with an accuracy rate of between 87.5 and 91 percent
In lab tests of the current NutriRay3D prototype, it estimated nutritional content of various foods with an accuracy rate of between 87.5 and 91 percent

For multi-ingredient dishes such as soups, ingredient lists for popular types are already in the database. If users want, however, they can manually enter the ingredients for recipes that they often prepare.

In lab tests of the current prototype, it estimated nutritional content of various foods with an accuracy rate of between 87.5 and 91 percent. While that may not be perfect, it's claimed to be significantly better than what's managed by people who simply self-report using paper logs or other traditional methods.

A team of U Washington scientists and students has now launched an Indiegogo project, to raise funds for commercialization of the technology. A pledge of US$240 will currently get you a NutriRay3D module, when and if they reach production.

Sources: University of Washington, Indiegogo

View gallery - 3 images
No comments
0 comments
There are no comments. Be the first!