By Dave DeFusco

Dr. Honggang Wang, chair of the Katz School’s Department of Computer Science and Engineering, has received a $600,000 grant from the National Institutes of Health to create an artificial intelligence platform that would recognize patterns in longitudinal dietary data.

The Innovative Pattern Analysis Tool, called iPAT, would employ a new machine-learning algorithm to facilitate comparisons between individual and population-level dietary patterns, and would generate evidence for dietary guidelines.

The pattern-recognition algorithm will use visualization techniques, such as charts, graphs or other visual representations, to make patterns in trajectories more apparent and interpretable, aiding researchers in understanding the data and forming hypotheses. A key component of the algorithm would be a validation process to ensure the accuracy and reliability of the identified patterns.

“The tool will provide a multi-view and comprehensive understanding of diet-quality trajectory patterns for various chronic disease outcomes,” said Dr. Wang. “It aims to uncover patterns that may not be apparent when analyzing individual studies in isolation, offering insights at different levels of granularity.”

Longitudinal dietary data is crucial for advancing understanding of the complex interactions between diet, health and various environmental and lifestyle factors. It provides insights that are essential for designing effective interventions and informing public health strategies related to nutrition.

Dr. Wang’s project, led by UMass researchers, is called “iPAT: Intelligent Diet Quality Pattern Analysis for Harmonized MA-National Trials,” which leverages data from seven studies funded by the National Institute of Diabetes and Digestive and Kidney Diseases; National Heart, Lung, and Blood Institute; and National Institute of Mental Health. It includes four longitudinal randomized controlled trials in Massachusetts and three large-scale longitudinal multisite national studies from the Women's Health Initiative and the Coronary Artery Risk Development in Young Adults study.

“We expect the evidence-based iPAT tool to contribute to the infrastructure for diet-related studies,” he said. “It will aid in the validation of evidence for dietary guidelines and support comparisons of individual dietary behavior with local and national diet-quality patterns.”

Dr. Wang will build on published research on data harmonization that he co-authored with researchers from the UMass Dartmouth, UMass Medical School, University of Minnesota, Fred Hutchinson Cancer Research Center, University of Alabama at Birmingham and Harvard T.H. Chan School of Public Health.

The researchers intend to create an accessible and expandable harmonized dietary database, and iPAT will be made available as an open-access tool for diet-related studies, fostering collaboration and knowledge-sharing. The harmonized data-driven approach will enhance researchers’ ability to address complex questions related to culture, age, gender and geographic variations in diet-quality patterns.

“The project aims to explore how diet quality may vary with context and time,” said Dr. Wang. “It holds broader implications for supporting adaptive interventions and enhancing the understanding of diet-related health risks.”