Abstract Body

Background: Accurate carbohydrate counting is an important aspect of diabetes management, but it can be challenging. iSpy is a novel mobile application designed to assist with carbohydrate counting by using machine learning to identify foods through images. Our objective was to evaluate iSpy’s usability and potential impact on carbohydrate counting accuracy.

Methods: For usability testing, three iterative cycles were conducted involving a total of 16 individuals (aged 8.5-17.0 years) with type 1 diabetes. Participants used iSpy to complete tasks while thinking aloud. Errors were noted, acceptability was assessed, and refinements were made before moving on to the next cycle. Next, iSpy was evaluated in a pilot randomized controlled trial with 22 iSpy users and 22 usual care controls aged 10-17 years. Primary outcome was change in carbohydrate counting ability over 3 months. Secondary outcomes included engagement, acceptability, and change in HbA_1c level.

Results: After 3 cycles of usability testing, no errors occurred that prevented a user from completing a task. For the pilot RCT, use of iSpy was associated with improved carbohydrate counting accuracy (total grams per meal, P=.008), reduced frequency of individual counting errors greater than 10 g (P=.047), and lower HbA_1c levels (P=.03). Qualitative interviews and acceptability scale scores were positive. No major technical challenges occurred. Moreover, 43% (9/21) of iSpy participants were still engaged, with usage at least once every 2 weeks, at the end of the study.

Conclusions: Our results provide evidence of efficacy and acceptability of a novel carbohydrate counting application, supporting the advancement of digital health apps for diabetes care among youth with type 1 diabetes. Further testing is needed, but iSpy may be a useful adjunct to traditional diabetes management.

Abstract Body

Accurate estimation of carbohydrates as well as other nutrients, are beneficial in the management of diabetes mellitus, leading to improved glycemic control and balanced dietary patterns. This presentation includes an overview of studies that compare conventional versus innovative dietary assessment methods and discusses the challenges of both categories. Various conventional dietary assessment methods exist (e.g. 24-h food recall, food frequency questionnaires) which aim at measuring dietary intake. However, existing methods encounter different drawbacks such as lack of precision, inaccuracy in portion size estimation and non-real-time feedback. Innovative methods of dietary assessment that have been introduced will be presented, with a focus on image-based apps which use food photos/videos as an input and ,via artificial intelligence, translate them into nutrients. Finally, based on international surveys, the preferences, criteria and barriers related to recommendation or usage of nutrition apps from the perspective of healthcare professionals and end-users will be presented.