AI-assisted dietary assessment and personalised “nudges” in urban youth: Feasibility and effectiveness of a mobile phone-based intervention to improve diets in Ghana

Non-PR
IFPRI1; DCA
Nutrition, Diets, and Health (NDH); Food and Nutrition Policy
Diet-related risks cause 20% of global deaths (~11 million) per year (Afshin et al., 2019). Increased urbanization in low- and middle-income countries accompanied by a nutrition transition, involving increased consumption of unhealthy, processed foods and reductions in physical activity, has exacerbated diet-related risks, leading to increases in rates of overweight and obesity (Popkin et al., 2020). Diets in urban settings are also affected by the reduced time for food preparation, convenience and affordability of unhealthy foods. We developed an artificial-intelligence-based phone application called Food Recognition Assistance and Nudging Insights (FRANI) to tackle these problems. FRANI was previously validated against observed weighed records, the gold standard for diet assessment, and shown to be as accurate as a trained dietician undertaking a standard, multi-pass 24-hr recall (Folson et al., 2023). FRANI was developed to also provide gamified nudges and personalized feedback designed to promote healthy food choices. The objectives of this study were to measure the acceptability, adherence, and likeability of FRANI, as well as its effects on the food choices of female youth in Accra, Ghana.