Using technology to improve dietary assessment methods

Dietary intake provides some of the most valuable insights for mounting intervention programs for prevention of chronic diseases and overweight. However, accurate assessment of diet is problematic due to self-reporting errors and respondent burden. Studies among adolescents suggest that innovative use of technology may improve the accuracy of diet information (Boushey, et al Euro J Clin Nutr 2009). The goal of the Technology Assisted Dietary Assessment (TADA) team is to develop, implement, and evaluate a mobile device (e.g., mobile phone, hand-held tablet) food record that will translate to an accurate account of daily food and nutrient intake. The mobile device provides a unique vehicle for collecting dietary information that reduces burden on record keepers. Images captured by users before and after foods are eaten can be used to estimate the amount of food and nutrients consumed using image analysis, visualization methods, and a nutrient database. Image classification accuracy of foods under good lighting conditions was 71% compared to 35% under bad lighting and 94% for replica foods (Mariappan, et al IS&T/SPIE, 2009). To aid with evidence based development, a total of 78 adolescents (11-18 y) and 57 adults (21-65 y) were recruited to participate in meal sessions. Participants were instructed to capture images with a mobile phone of all foods and beverages, and the fiducial marker (an item of known size), before and after eating. Results of these studies have been used to test the system and improve the interaction design, thus starting a new era for diet assessment with potentially improved participant cooperation and accuracy.