Dataset Information

PROTEIN AI Advisor: A Knowledge-Based Recommendation Framework Using Expert-Validated Meals for Healthy Diets.

ABSTRACT: AI-based software applications for personalized nutrition have recently gained increasing attention to help users follow a healthy lifestyle. In this paper, we present a knowledge-based recommendation framework that exploits an explicit dataset of expert-validated meals to offer highly accurate diet plans spanning across ten user groups of both healthy subjects and participants with health conditions. The proposed advisor is built on a novel architecture that includes (a) a qualitative layer for verifying ingredient appropriateness, and (b) a quantitative layer for synthesizing meal plans. The first layer is implemented as an expert system for fuzzy inference relying on an ontology of rules acquired by experts in Nutrition, while the second layer as an optimization method for generating daily meal plans based on target nutrient values and ranges. The system's effectiveness is evaluated through extensive experiments for establishing meal and meal plan appropriateness, meal variety, as well as system capacity for recommending meal plans. Evaluations involved synthetic data, including the generation of 3000 virtual user profiles and their weekly meal plans. Results reveal a high precision and recall for recommending appropriate ingredients in most user categories, while the meal plan generator achieved a total recommendation accuracy of 92% for all nutrient recommendations.

SUBMITTER: Stefanidis K

PROVIDER: S-EPMC9612332 | biostudies-literature | 2022 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

PROTEIN AI Advisor: A Knowledge-Based Recommendation Framework Using Expert-Validated Meals for Healthy Diets.

Stefanidis Kiriakos K Tsatsou Dorothea D Konstantinidis Dimitrios D Gymnopoulos Lazaros L Daras Petros P Wilson-Barnes Saskia S Hart Kathryn K Cornelissen Véronique V Decorte Elise E Lalama Elena E Pfeiffer Andreas A Hassapidou Maria M Pagkalos Ioannis I Argiriou Anagnostis A Rouskas Konstantinos K Hadjidimitriou Stelios S Charisis Vasileios V Dias Sofia Balula SB Diniz José Alves JA Telo Gonçalo G Silva Hugo H Bensenousi Alex A Dimitropoulos Kosmas K

Nutrients 20221021 20

AI-based software applications for personalized nutrition have recently gained increasing attention to help users follow a healthy lifestyle. In this paper, we present a knowledge-based recommendation framework that exploits an explicit dataset of expert-validated meals to offer highly accurate diet plans spanning across ten user groups of both healthy subjects and participants with health conditions. The proposed advisor is built on a novel architecture that includes (a) a qualitative layer for ...[more]

PMID: 36297118

Similar Datasets

Project description:BackgroundCombinations of foods that provide suitable levels of nutrients and energy are required for optimum health. Currently, however, it is difficult to define numerically what are 'suitable levels'.ObjectiveTo develop new metrics based on energy considerations-the Nutrient Balance Concept (NBC)-for assessing overall nutrition quality when combining foods and meals.MethodThe NBC was developed using the USDA Food Composition Database (Release 27) and illustrated with their MyPlate 7-day sample menus for a 2000 calorie food pattern. The NBC concept is centered on three specific metrics for a given food, meal or diet-a Qualifying Index (QI), a Disqualifying Index (DI) and a Nutrient Balance (NB). The QI and DI were determined, respectively, from the content of 27 essential nutrients and 6 nutrients associated with negative health outcomes. The third metric, the Nutrient Balance (NB), was derived from the Qualifying Index (QI) and provided key information on the relative content of qualifying nutrients in the food. Because the Qualifying and Disqualifying Indices (QI and DI) were standardized to energy content, both become constants for a given food/meal/diet and a particular consumer age group, making it possible to develop algorithms for predicting nutrition quality when combining different foods.ResultsCombining different foods into composite meals and daily diets led to improved nutrition quality as seen by QI values closer to unity (indicating nutrient density was better equilibrated with energy density), DI values below 1.0 (denoting an acceptable level of consumption of disqualifying nutrients) and increased NB values (signifying complementarity of foods and better provision of qualifying nutrients).ConclusionThe Nutrient Balance Concept (NBC) represents a new approach to nutrient profiling and the first step in the progression from the nutrient evaluation of individual foods to that of multiple foods in the context of meals and total diets.

Dataset Information

PROTEIN AI Advisor: A Knowledge-Based Recommendation Framework Using Expert-Validated Meals for Healthy Diets.

Publications

PROTEIN AI Advisor: A Knowledge-Based Recommendation Framework Using Expert-Validated Meals for Healthy Diets.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets