Research behind TrueFood

[1] Prevalence of processed foods in major US grocery stores. [Nature Food, 2025]

[2] Machine learning prediction of the degree of food processing. [Nature Communications, 2023]

[3] Nutrient concentrations in food display universal behaviour. [Nature Food, 2022]

Limitations

The machine learning algorithm behind TrueFood, FoodProX (FPro), partially draws from expertise-based food processing classifications from NOVA due to the lack of data regarding compound concentrations, food matrix alterations, or industrial processing techniques. FPro functions as a quantitative algorithm, utilizing standardized inputs to generate reproducible and continuous scores that enhances reliability, transparency, and interpretability. While FPro reduces errors from the descriptive nature of manual classifications by nutrition specialists, it does not eliminate them since FPro is trained on expertise-based systems.

The chemical composition of branded products is partially captured by the nutrition facts table and partially reported in the ingredient list. However, comprehensive and internationally well-regulated data on food ingredients is limited. Therefore, FPro focuses on the nutrition facts alone which enhance reproducibility and show excellent performance in determining a food's NOVA class. The few nutrients available on food packaging increase the risk of identifying products with similar nutrition facts but distinct food matrices (e.g., pre-frying, puffing, extrusion-cooking). Indeed, if the nutrient panel fails to capture food matrix alterations, then FPro and the substitution algorithm implemented on TrueFood will remain blind to these chemical-physical changes.

Other Relevant References

[4] Consumption of ultra-processed foods and cancer risk: results from NutriNet-Santé prospective cohort.

[5] Prospective association between ultra-processed food consumption and incident depressive symptoms in the French NutriNet-Santé cohort.

[6] Association between ultraprocessed food intake and cardiovascular health in US adults: a cross-sectional analysis of the NHANES 2011–2016.

[7] Ultra-processed food consumption and type 2 diabetes incidence: A prospective cohort study.

[8] Ultra-Processed Diets Cause Excess Calorie Intake and Weight Gain: An Inpatient Randomized Controlled Trial of Ad Libitum Food Intake.

[9] Association between dietary contribution of ultra-processed foods and urinary concentrations of phthalates and bisphenol in a nationally representative sample of the US population aged 6 years and older.

[10] Processed food—an experiment that failed.

[11] Profits and pandemics: prevention of harmful effects of tobacco, alcohol, and ultra-processed food and drink industries.

[12] Ultra-processed foods: what they are and how to identify them.

[13] The UN Decade of Nutrition, the NOVA food classification and the trouble with ultra-processing.

[14] Global trends in ultraprocessed food and drink product sales and their association with adult body mass index trajectories.

[15] The Healthfulness of the US Packaged Food and Beverage Supply: A Cross-Sectional Study.

[16] A Fit-for-Purpose Nutrient Profiling Model to Underpin Food and Nutrition Policies in South Africa.