AI calorie-tracking apps miss up to 345 calories per meal, NIH study reveals

Your AI calorie-tracking app may be off by 345 calories

In a test of four photo-based apps, researchers found that calorie and fat estimates were about one-third too low on average. How AI Estimates Calories From Food Photos Photo-based calorie tracking relies on AI image recognition to identify the foods shown in a picture and estimate the size of each portion. The app then compares those estimates with nutrition databases to calculate calories and other nutrients. "Photo-based calorie tracking apps are very popular, especially for people trying to manage their health or lose weight," said Aaron Hengist, a postdoctoral visiting fellow with the Intramural Program of the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), part of the National Institutes of Health. "However, the accuracy of many of these apps has not been thoroughly evaluated. Our study helps address this question by looking at whether these apps can reliably estimate calories." Olivia Charles, a postbaccalaureate intramural research training fellow at NIDDK, presented the findings at NUTRITION 2026, the flagship annual meeting of the American Society for Nutrition, held July 25-28 in National Harbor, Maryland, just outside Washington, D.C. Testing Apps With Precisely Measured Meals The project is part of a broader nutrition study at the NIH Clinical Center that is investigating how the body processes nutrients on either a low-carbohydrate (ketogenic) diet or a standard diet. Meals used in the clinical trial are prepared in a controlled metabolic kitchen, where researchers measure ingredients to the nearest 0.1 gram. This gave the team a highly accurate reference for evaluating the apps. Researchers collected standardized photographs of 102 meals prepared for the diet study. They then submitted the images to MyFitnessPal, LoseIt!, CalAI and Appediet to see how closely each app's estimates matched the known nutritional content. "By using meals prepared in a tightly controlled metabolic kitchen, we were able to compare the apps' estimates against a precise reference," said Hengist. "This kind of direct, high-quality comparison hasn't been available before." Apps Missed Hundreds of Calories Across all four apps, estimated calorie totals were about 250 to 345 calories too low per meal on average. The apps also underestimated fat by approximately 30 grams. MyFitnessPal and LoseIt! were more accurate when analyzing higher-calorie meals than they were with lower-calorie meals. All four apps also produced more consistent estimates for carbohydrates than for other macronutrients. "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," said Hengist. "These apps tend to underestimate calories, especially from fats, so what they actually ate is likely higher than what the app shows." Keto Meals May Be Harder for AI to Measure Following the first analysis, the researchers tested more than 200 additional meals to investigate which factors might influence app accuracy. Preliminary findings indicate that the apps may have greater difficulty evaluating meals from a low-carb ketogenic diet. These meals often contain more fat, which the apps tended to underestimate consistently. The researchers suggest that combining photo-based tools with traditional methods of evaluating food intake and diet quality could make calorie tracking more accurate in everyday use. Charles presented this research on Saturday, July 25, during the President's Oral Session in the Gaylord National Resort & Convention Center (abstract). Abstracts presented at NUTRITION 2026 were reviewed and selected by a committee of experts. However, they have generally not completed the full peer review process required for publication in a scientific journal. The results should therefore be considered preliminary until they appear in a peer-reviewed publication.

Photo-Based Calorie-Tracking Apps May Underestimate Energy in Meals | Newswise

Study examining four AI-powered apps found that calories and fat content were underestimated by roughly one-third Newswise -- National Harbor, Md. (July 25, 2026) -- Thanks to advances in AI, popular calorie-tracking apps can now use a photo of a meal to estimate the nutritional content of what you're eating. Although these apps are convenient, a new study found that four photo-based apps underestimated the calories and fat content of meals by about a third. Photo-based calorie tracking uses AI image recognition to identify foods in a meal photo and estimate portion sizes. This information is then compared with nutrition databases to calculate estimated calories and nutrients. "Photo-based calorie tracking apps are very popular, especially for people trying to manage their health or lose weight," said Aaron Hengist, a postdoctoral visiting fellow with the Intramural Program of the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), part of the National Institutes of Health. "However, the accuracy of many of these apps has not been thoroughly evaluated. Our study helps address this question by looking at whether these apps can reliably estimate calories." Olivia Charles, a postbaccalaureate intramural research training fellow at NIDDK, will present the findings at NUTRITION 2026, the flagship annual meeting of the American Society for Nutrition, held July 25-28 in National Harbor, Maryland, just outside Washington, D.C. The new research is part of a larger diet study at the NIH Clinical Center examining how the body processes nutrients when a person eats a low-carbohydrate (ketogenic) or standard diet. Meals for this clinical trial are prepared in a controlled metabolic kitchen, a type of research kitchen where ingredients are carefully weighed to the nearest 0.1 gram. To evaluate how well photo-based calorie-tracking apps estimate nutritional content from images, the researchers acquired standardized photos of 102 meals prepared as part of the diet study and then ran them through the MyFitnessPal, LoseIt!, CalAI and Appediet apps. "By using meals prepared in a tightly controlled metabolic kitchen, we were able to compare the apps' estimates against a precise reference," said Hengist. "This kind of direct, high-quality comparison hasn't been available before." All four apps underestimated the calorie content of meals by about 250 to 345 calories, on average, and fat by about 30 grams. The analysis also showed that the MyFitnessPal and LoseIt! apps estimated the energy content of higher-calorie meals more accurately than lower-calorie meals and that all four apps estimated carbohydrates more consistently than other macronutrients. "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," said Hengist. "These apps tend to underestimate calories, especially from fats, so what they actually ate is likely higher than what the app shows." After performing the initial analysis, the researchers examined more than 200 additional meals to better understand what might affect the accuracy of the apps. Their early results suggest that the apps struggle more with meals that are part of a low-carb ketogenic diet, likely because the higher fat content of these meals tends to be consistently underestimated. The investigators say that combining photo-based features with more traditional methods of measuring diet quality could improve the real-world accuracy of calorie-tracking apps. Charles will present this research at 4:42 - 4:54 p.m. EDT on Saturday, July 25, during the President's Oral Session in the Gaylord National Resort & Convention Center (abstract; presentation details). Please note that abstracts presented at NUTRITION 2026 were evaluated and selected by a committee of experts but have not generally undergone the same peer review process required for publication in a scientific journal. As such, the findings presented should be considered preliminary until a peer-reviewed publication is available. Images available. About NUTRITION 2026 NUTRITION 2026 is the flagship meeting of the American Society for Nutrition and the premier educational event for nutritional professionals around the globe. NUTRITION brings together lab scientists, practicing clinicians, population health researchers and community intervention investigators to identify solutions to today's greatest nutrition challenges. The meeting also attracts rising leaders in the field, including undergraduate, graduate and medical students. NUTRITION 2026 will be held July 25-28 in National Harbor, Maryland, just outside Washington, D.C. https://nutrition.org/meeting/ #Nutrition2026 About the American Society for Nutrition (ASN) ASN is the preeminent professional organization for nutrition research scientists and clinicians around the world. Founded in 1928, the society brings together the top nutrition researchers, medical practitioners, policy makers and industry leaders to advance knowledge and application of nutrition. ASN publishes four peer-reviewed journals and provides education and professional development opportunities to advance nutrition research, practice, and education. Since 2018, the American Society for Nutrition has presented NUTRITION, the leading global annual meeting for nutrition professionals. http://www.nutrition.org Find more news briefs from NUTRITION 2026 at: https://www.eurekalert.org/newsroom/nutrition2026.