A Novel Approach for Determining Meal Plan for Gestational Diabetes Mellitus Using Artificial Intelligence

Abstract

Estimating energy expenditure and meal plan plays important roles in the treatment of gestational diabetes mellitus, which is defined as any degree of glucose intolerance with onset or first recognition during pregnancy. Some approaches have been proposed; however, they have limitations including high cost, relative complexity, trained personnel requirements or locality. In this study, we propose an approach for estimating the energy expenditure and meal plan by using artificial intelligence. The proposed approach consists of three main stages including energy expenditure estimation, macronutrient intake estimation and meal plan determination. The neural network is used to estimate the energy expenditure, and then the meal plan is determined by using the genetic algorithm (GA), which is a popular method for solving optimization problems based on natural selection and genetics. The fitness function with penalty was used in GA to deal with constraint problems. The proposed method can obtain the root mean square error and mean absolute percentage error of 15.23 ± 7.4 kcal and 1 ± 0.8%, respectively. The Pearson correlation coefficient, which measures the strength of the association between the two measurements, was 0.99. In meal plan determination, the results from GA agreed with those from nutritionists. The Pearson correlation coefficient was 0.95. The energy expenditure and meal plan are determined by soft computing with flexible ways. They can adapt to particular regions or group of patients.

Meal Planning for Alzheimer's Disease Using an Ontology-Assisted Multiple Criteria Decision-Making Approach

“…In the proposed algorithm, the meal-planning problem is split into well-defined sub-problems, and a multi-level structure is employed to solve the problem. Further, Huynh and Hoang (2022) propose an AI-based meal planning approach by estimating the energy expenditure of a user. The proposed approach first estimates a user's energy expenditure using a neural network.…”