AraDQ

AraDQ quantifies disease symptoms in flood-inoculated Arabidopsis seedlings using deep neural networks for object detection and segmentation.

Key Features:

• Deep neural network-based image analysis: Utilizes deep neural networks for object detection and segmentation of Arabidopsis seedlings.
• Color calibration: Requires a fixed-size 24-color balance card placed next to objects of interest on a white background for accurate and reproducible results regardless of image acquisition method.
• Automated parameter calculation: Automatically calculates 10 different color and morphological parameters for individual seedlings in a given image.
• Temporal comparison: Enables assessment of disease-associated phenotypic changes by comparing plant images captured before and after infection.

Scientific Applications:

• Disease severity quantification: Provides image-based measurement of disease symptoms as an alternative to destructive pathogen growth assays and transcript analysis of stress-responsive genes.
• Mutant phenotyping: Captures subtle changes in plant color and morphology with high sensitivity for bacterial and plant mutants with altered virulence and disease resistance.
• Large-scale phenotyping studies: Enables real-time, non-destructive, automated measurement suitable for high-throughput disease susceptibility analysis.

Methodology:

The image processing pipeline uses deep neural networks for object detection and segmentation, with color calibration via a fixed-size 24-color balance card, and calculates 10 color and morphological parameters per seedling.