Self-reported dietary assessments are prone to measurement errors, affecting nutrient profile accuracy. This study introduces METRIC, a deep-learning approach leveraging gut microbial compositions to correct random errors in self-reported dietary assessments. METRIC demonstrates excellent performance in minimizing simulated random errors, particularly for nutrients metabolized by gut bacteria, across synthetic and real-world datasets. Even without gut microbial compositions, METRIC shows good correction performance. Further research is needed to assess METRIC's utility in correcting actual measurement errors.
Publisher
Nature Communications
Published On
Oct 22, 2024
Authors
Tong Wang, Yuanqing Fu, Menglei Shuai, Ju-Sheng Zheng, Lu Zhu, Andrew T. Chan, Qi Sun, Frank B. Hu, Scott T. Weiss, Yang-Yu Liu
Tags
dietary assessments
measurement errors
deep learning
gut microbiome
nutrient accuracy
correction methods
random errors
Related Publications
Explore these studies to deepen your understanding of the subject.
