Search for a command to run...
<sec> <title>BACKGROUND</title> Thorough dietary assessment is essential to obtain accurate food and nutrient intake data, yet challenging due to limitations of current methods. Image-based methods may decrease energy underreporting and increase validity of self-reported dietary intake. Keenoa is an image-assisted food diary that integrates artificial-intelligence food recognition. We hypothesized that Keenoa is as valid for dietary assessment as the Automated Self-Assessment (ASA) 24-Canada and better appreciated by users. </sec> <sec> <title>OBJECTIVE</title> to evaluate the relative validity of Keenoa against a 24-hour validated web-based food recall platform (ASA24) in both healthy individuals and those living with diabetes. Secondary objectives were to compare the proportion of under and over-reporters between tools, and to assess the user’s appreciation of the tools. </sec> <sec> <title>METHODS</title> Using a randomized crossover design, participants completed 4 days of Keenoa food tracking and 4 days of ASA24 food recalls. The System Usability Scale (SUS) assessed perceived ease of use. Differences in reported intakes were analyzed using paired t-tests or Wilcoxon signed-rank test and deattenuated correlations, by Spearman’s coefficient. Agreement and bias were determined using Bland-Altman’s test. Weighted Cohen’s kappa was used for cross-classification analysis. Energy underreporting was defined as a ratio of reported energy intake:estimated resting energy expenditure <0.9. </sec> <sec> <title>RESULTS</title> One hundred and thirty-six participants were included (46.1 ± 14.6 years; 36% men; 23% with diabetes). Mean (± SD) reported energy intakes (kcal/d) were, in men, 2171 ± 553 with Keenoa and 2118 ± 566 with ASA24 (P=.38), and in women, 1804 ± 404 with Keenoa and 1784 ± 389 with ASA 24 (P=0.61). The overall mean difference (kcal/d) was -32 (95%CI: -97 to 33), limit of agreement of -789 to 725, indicating acceptable agreement between tools, without bias. Mean reported macronutrient, calcium, potassium, and folate intakes did not significantly differ between tools. Reported fiber and iron intakes were higher, and sodium intake lower, with Keenoa than ASA24. Intakes in all macronutrients (r=0.48 to 0.73) and micronutrients analyzed (r=0.40 to 0.74) correlated (all P<.05) between tools. Weighted Cohen’s kappa scores ranged from 0.30-0.52 (all P<.001). Under-reporting rate was of 8.8% with both tools. Mean SUS scores were higher for Keenoa than ASA24 (77 vs. 53/100, P<.001); 75% of participants preferred Keenoa. </sec> <sec> <title>CONCLUSIONS</title> The Keenoa application showed moderate to strong relative validity against ASA24 for energy, macronutrient, and most micronutrient intakes analyzed in healthy adults and those with diabetes. Keenoa is a new, alternative tool that may facilitate the work of dietitians and nutrition researchers. The perceived ease of use may improve food tracking adherence over longer periods. </sec> <sec> <title>CLINICALTRIAL</title> This study was registered on the Dietary Assessment Calibration/Validation (DACV) Register from the National Cancer Institute (NIH). </sec>