Database Shows Degree of Processing for More Than 50,000 Food Items

MONDAY, Jan. 13, 2025 (HealthDay News) — A new database, GroceryDB, can show the degree of processing of more than 50,000 food items sold at Walmart, Target, and Whole Foods, according to a study published online Jan. 13 in Nature Food.

Noting that information on the degree of processing characterizing an item in a store is not straightforward to obtain, Babak Ravandi, Ph.D., from Northeastern University in Boston, and colleagues built on the versatility and scalability of the previously developed food processing score (FPro), which harnessed machine learning, to develop Grocery DB, a database with more than 50,000 food items from major U.S. grocery store websites. GroceryDB has the ability to show the degree of processing of food items and potential alternatives in the surrounding food environment.

For each food, the extent of food processing was determined using FPro; Grocery DB then assigned an FPro score to all foods from the included grocery stores by leveraging a machine learning classifier, which takes mandatory information from nutrition labels as input. The authors note that the extensive data gathered on ingredient lists and nutrition facts allowed for a large-scale analysis of ingredient patterns and degree of food processing, which were categorized by store, food category, and price range. The data also allowed for quantification of the individual contributions of more than 1,000 ingredients to ultraprocessing.

“People can use this information, but our goal would be to push this to become a large-scale, data-driven tool to improve public health,” coauthor Giulia Menichetti, Ph.D., also from Northeastern University, said in a statement. “Most research activities in nutrition still depend on manual curation, but our study shows that artificial intelligence and data science can be used to scale up.”

One author disclosed ties to the health technology industry.

Abstract/Full Text (subscription or payment may be required)