Location: Wheat Health, Genetics, and Quality Research
Project Number: 2090-21000-039-022-S
Project Type: Non-Assistance Cooperative Agreement
Start Date: Sep 1, 2026
End Date: Feb 28, 2028
Objective:
The USDA-ARS Wheat Health, Genetics, and Quality Research Unit and the University of Delaware are collaborating to develop a robust and objective AI based phenotyping framework for quantifying root rot severity caused by soilborne pathogens, including Pythium species, in maize and wheat. The resulting framework will provide standardized and objective disease phenotyping across environments, accelerating genetic studies and breeding efforts aimed at improving root rot resistance in maize, wheat, and other cereal crops.
Approach:
The approach leverages root imaging data and advanced artificial intelligence methods, including Vision Foundation Models (VFMs), combined with multimodal imaging technologies to enable accurate, scalable, and high-throughput assessment of root rot severity. Ground-truth disease measurements will be established using both traditional visual disease ratings and objective pathogen quantification by qPCR. Model development, training, and deployment will leverage USDA SCINet computational resources to support large-scale image analysis and AI model optimization.