ASA24
ASA24 automates collection and coding of self-reported 24-hour dietary recalls to estimate nutrient intake for nutritional epidemiology and dietary assessment research.
Key Features:
- Automatic coding: Automatically matches reported foods to standardized food items and nutrient codes by referencing the ASA24 Food and Nutrient Database.
- ASA24 Food and Nutrient Database: Maps foods to nutrient values using a database derived from the USDA National Nutrient Database for Standard Reference.
- Nutrient coverage: Produces a defined set of nutrient outputs that include fewer nutrient variables than the Nutrition Data System for Research (NDSR) and does not cover all nutrients in the NCC Food and Nutrient Database.
- Computational augmentation: Compatible with downstream machine learning and database-matching approaches, including XGB-Regressor and the "Nutrient + Text" matching algorithm, to estimate nutrients not present in ASA24 outputs (e.g., lactose).
- 24-hour dietary recall data: Captures self-reported intake over a 24-hour period for quantitative nutrient estimation.
Scientific Applications:
- Nutritional epidemiology: Estimating nutrient intake distributions for population- and cohort-based diet-disease research.
- Diet-disease relationship studies: Supporting analyses that relate reported dietary intake to health outcomes.
- Nutritional intake analysis: Providing nutrient-level data for analyses of dietary patterns and exposures.
- Public health nutrition surveillance: Enabling large-scale dietary assessment for population monitoring and surveillance.
- Augmented nutrient estimation: Allowing researchers to apply ML and database-matching methods to estimate nutrients absent from ASA24 outputs for targeted nutrient investigations.
Methodology:
Reported foods are automatically matched to entries in the ASA24 Food and Nutrient Database (derived from the USDA National Nutrient Database for Standard Reference); researchers have applied machine learning models (e.g., XGB-Regressor) and database-matching algorithms (e.g., "Nutrient + Text") to estimate NCC-exclusive nutrients such as lactose from ASA24 outputs.
Topics
Details
- Tool Type:
- web application
- Added:
- 1/14/2020
- Last Updated:
- 11/24/2024
Operations
Publications
Chin EL, Simmons G, Bouzid YY, Kan A, Burnett DJ, Tagkopoulos I, Lemay DG. Nutrient Estimation from 24-Hour Food Recalls Using Machine Learning and Database Mapping: A Case Study with Lactose. Nutrients. 2019;11(12):3045. doi:10.3390/nu11123045. PMID:31847188. PMCID:PMC6950225.