BACKGROUND
Due to its influence on health, access to a good diet is one of the great objectives of the health services in many countries. There are many electronic services that offer advice on diet, recommending specific foods or recipes. Additionally, as information processing techniques advance, options to tackle a multi-factorial and individualized nutritional recommendation also increase in number, allowing the recommendation of complete menus taking into account several parameters.
OBJECTIVE
In this article we present and validate a personalized nutrition system based on an application (APP) for smart devices with the capacity to offer an adaptable menu to the user.
METHODS
The APP was developed following a structured recommendation generation scheme, where the characteristics of the menus of 20 users were evaluated. From these, a user profile was developed with their nutritional requirements and menus were generated for 2 weeks. These menus were evaluated by comparing their mean nutritional content versus the mean nutrient composition retrieved from dietary records.
RESULTS
The generated menus showed great similarity to those obtained from the user dietary questionnaires. However, the generated menus showed less variability regarding the amounts of micronutrients reached. This lower variability was accompanied by higher quantities of most of the micronutrients tested in the user menus. The macronutrient deviations were also corrected in the generated menus, offering a better adaptation to the biometric parameters provided by the users.
CONCLUSIONS
The presented system is a good tool for the generation of menus that are adapted to the user characteristics. It is therefore a starting point to carry out nutritional interventions where it is necessary to control the amount of nutrients. These nutritional interventions will be important to validate other modules of the system.
CLINICALTRIAL
ISRCTN63745549
INTERNATIONAL REGISTERED REPORT
RR2-https:// doi.org/10.3390/foods11101480