This paper introduces FoodProX, a machine learning algorithm that predicts the degree of food processing. It finds that over 73% of the US food supply is ultra-processed and demonstrates a correlation between ultra-processed food consumption and increased risk of metabolic syndrome, diabetes, angina, elevated blood pressure, biological age, and reduced vitamin bioavailability. Replacing ultra-processed foods with less processed alternatives shows significant health improvements, highlighting the potential of accessible processing information to enhance population health.
