Automated knowledge-based nutrition health assessment, recommendation, progress evaluation, and meal planning. (c2018)

Abstract

Establishing a healthy lifestyle has become a very important aspect in people’s lives. The latter requires maintaining a healthy nutrition by considering the nature and quantity of foods being consumed, as well as maintaining an active lifestyle including the necessary amount of physical exercise to regulate one’s intake and consumption of calories and nutrients. As a result, people reach out for nutrition experts which entails various obstacles: (i) the cost of seeking an expert’s help which is recurring and non-trivial, (ii) the time commitment required from a person to attend regular meetings with the expert, and (iii) the need for readily accessible health services which might be difficult to provide by a human expert. In this master thesis study, we design, implement, and evaluate a novel framework titled PIN: a computerized solution for Personalized Intelligent Nutrition recommendations. PIN consists of four main modules allowing four essential complementary functionalities: (i) Weight Assessment and Recommendation (WAR), (ii) Caloric Intake and Exercise Recommendation (CIER), (iii) Progress Evaluation and Recommendation Adjustment (PERA), and (iv) personalized Meal Plan Generation (MPG) and adaptation following patient chosen parameters (e.g., food preference, food compatibility, price, etc.). While most existing computerized solutions focus solely on the meal plan generation task, PIN provides the first full-fledged solution for nutrition health assessment, which results are required to run the meal planning task. It relies on the fuzzy logic paradigm to simulate human expert health assessment including weight, caloric intake, and exercise recommendations as well as progress evaluation and recommendation adjustments. PIN also provides a novel contribution in meal planning, introducing an adaptation of the transportation optimization problem to dynamically generate, change, adapt, and self-evaluate meal plans following the patient’s needs, compared with most existing meal planning solutions which fail to integrate all the different essential factors (meal-food compatibility, inner-food compatibility, preferences, diversity, and variety) while producing meal plans. We have conducted a large battery of experiments involving 50 patient profiles, 11 nutrition expert evaluators, and 5 non-expert testers, to test the performance of PIN, evaluating its health assessment and meal plan generation quality. Results highlight PIN’s assessment and recommendation qualities which are on a par with and sometimes surpass those of human nutrition experts.