Validating an Inference Engine for Data-driven Personalized Diet Goals
Principal Investigator
Name
Marissa Burgermaster
Degrees
Ph.D., Behavioral Nutrition; M.A. Biomedical Informatics
Institution
University of Texas at Austin
Position Title
Assistant Professor
About this CDAS Project
Study
IDATA
(Learn more about this study)
Project ID
IDATA-46
Initial CDAS Request Approval
Jun 15, 2021
Title
Validating an Inference Engine for Data-driven Personalized Diet Goals
Summary
Our inference engine uses ASA24 diet recall data to compute a patient's status for each of nine previously developed, MyPlate-based goals (e.g., “Make half my grains whole”). The inference engine synthesizes a patient’s data and compares it to evidence-based targets for nutrient consumption personalized for patient characteristics (e.g., kcal intake/sex/age) and presents a prioritized list and rationale based on standardized status levels across the goals. The purpose of this project is to use ASA24 Recalls, participant characteristics, and energy expenditure data from the iData Dataset to ensure that the inference engine presents an appropriate range of goals across a large sample and for individual participants.
Aims
1. To describe the frequency of goal achievement, goal progress status, and consistency across the sample and individual respondents.
2. To assess the participant characteristics, anthropometry, and doubly labeled water energy expenditure related to inference engine recommendation of calorie intake.
Collaborators
Madalyn Rosenthal, BS – The University of Texas at Austin
Dagny Larson, RD – The University of Texas at Austin