TOPLINE

Photo-based meal-tracking apps underestimated calorie content by about 33% and fat content by about 30 g per meal compared with nutrition databases.

METHODOLOGY

• Photo-based calorie tracking uses AI image recognition to identify foods in a meal photo and estimate portion sizes; the information is then compared with nutrition databases to calculate estimated calories and nutrients.
• To evaluate the accuracy of the photograph-based AI features from four calorie-tracking apps, researchers acquired photos of 102 meals prepared in a controlled metabolic kitchen as part of a larger US National Institutes of Health diet study.
• The photos were analyzed using MyFitnessPal, Lose It!, Cal AI, and Appediet.
• The app-generated estimates were compared with values from national food composition tables.

TAKEAWAY

• All four apps underestimated energy content by an average of about 250-345 kcal per meal and fat content by an average of about 30 g per meal.
• Lose It! and Cal AI underestimated carbohydrates by about 14 g on average, whereas the other two apps showed no significant difference from national data.
• MyFitnessPal and Lose It! estimated energy content more accurately in higher-calorie meals than in lower-calorie meals.
• Researchers suggest that combining photo-based features with more traditional methods of measuring diet quality could improve the real-world accuracy of calorie-tracking apps.

IN PRACTICE

“People using a photo-based tracking app without adjusting the portions or entering the amounts of food should take the results with a grain of salt,” co-author Aaron Hengist, a postdoctoral visiting fellow at the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), said in a press release. “These apps tend to underestimate calories, especially from fats, so what they actually ate is likely higher than what the app shows.”

SOURCE

Co-author Olivia Charles, a postbaccalaureate intramural research training fellow at the NIDDK, presented the findings at NUTRITION 2026.

LIMITATIONS

The study was presented at a meeting and has not yet undergone peer review. More research is needed to confirm the results.

DISCLOSURES

The research was supported by the Intramural Research Program of the NIDDK within the US National Institutes of Health. The authors declared having no relevant financial relationships with ineligible companies.