Assessing Mediterranean Diet Adherence with the Smartphone: The Medipiatto Project

Abstract

The Mediterranean diet (MD) is regarded as a healthy eating pattern with beneficial effects both for the decrease of the risk for non-communicable diseases and also for body weight reduction. In the current manuscript, we propose an automated smartphone application which monitors and evaluates the user’s adherence to MD using images of the food and drinks that they consume. We define a set of rules for automatic adherence estimation, which focuses on the main MD food groups. We use a combination of a convolutional neural network (CNN) and a graph convolutional network to detect the types of foods and quantities from the users’ food images and the defined set of rules to evaluate the adherence to MD. Our experiments show that our system outperforms a basic CNN in terms of recognizing food items and estimating quantity and yields comparable results as experienced dietitians when it comes to overall MD adherence estimation. As the system is novel, these results are promising; however, there is room for improvement of the accuracy by gathering and training with more data and certain refinements can be performed such as re-defining the set of rules to also be able to be used for sub-groups of MD (e.g., vegetarian type of MD).

Weekend–Weekday Differences in Adherence to the Mediterranean Diet among Spanish University Students

“…While 24HR registers consumption of food groups from the day before, e-12HR allows users to register their consumption during the same day (normally with a maximum period of time between consumption of the food and it being registered of 12 h). Regarding the difficulty of estimating the size of the servings consumed, we must take into account that AMD assessment does not require a precise estimation of the size of the foods consumed, but rather a precise recognition of the food groups consumed together with an approximate estimation of the servings [ 51 ]. In this sense, e-12HR allows for collecting data on consumption for all the food groups needed to calculate AMD.…”

“…For food groups that have scores greater than 1, the scoring rules are as follows [ 51 ]: Fruits contribute 1 point for 1 to 2 servings, 2 points for 2 to 3 servings, and 3 points for 3–6 servings per day. Vegetables contribute 1 point for 2 to 4 servings, 2 points for 4 to 6 servings, and 3 points for ≥6 servings per day.…”

“…On top of this, the score on the AMD index was related with one of three levels of AMD [ 51 ]: low (0–8 points), moderate (9–15 points) or high (16–24 points).…”

“…However, it was necessary to make some modifications in order to adapt them to the characteristics of e-12HR (Table 1). For food groups that have scores greater than 1, the scoring rules are as follows [51]:…”

Mediterranean diet scoring systems: understanding the evolution and applications for Mediterranean and non-Mediterranean countries

“…The application of smartphone and other technology to assess dietary intake has recently been applied to the MedD (116) . The methodological intricacies and appropriate calculation of amount ingested require further validation studies (117) , but it is a promising new approach (116) .…”