For over a century, corn tar spot bided its time away from the U.S. That came to an end when the fungal disease was first reported and identified in Indiana and Illinois in 2015. Since then, tar spot has spread and posed a challenge to growers.
Where did the disease come from, how did it come from, and how different are the pathogens in other countries compared to the U.S.? “We probably don't have those populations in the U.S. yet, but they could be much more damaging than the populations that exist today,” says CD Cruz, an associate professor of botany and plant pathology at Purdue University.
Cruz and his colleagues are combining statistics, data science, epidemiology, microbiology, artificial intelligence, computer vision, and ongoing stakeholder feedback to better understand the dynamics of corn tar spot disease. Their research is funded by two grants totaling about $1.1 million from the U.S. Department of Agriculture’s National Institute of Food and Agriculture.
The implementation of these high-tech data collection and analysis methods will make critical information about corn pathology more widely and quickly available. “The integration of AI-based digital technologies and point-of-care diagnostics has the potential to provide faster and more accurate information to stakeholders,” says Cruz.
In previous work, Crews and his colleagues developed a digital method to quantify tar spot in corn. They tested the Stroma Contour Detection Algorithm (SCDA) against human raters under field conditions. They were able to quantify the trends, and their results were published in a paper published in the journal Frontiers in Plant Science in 2021.
Cantaloupe and other melon varieties currently attracting plant breeders' attention
Earlier this year, Crews' team ran similar experiments with SCDA 2.0, again with promising results, though inaccuracies remain in fully automated scenarios. “We are refining our training methodology to significantly improve the accuracy and performance of our models,” Crews added.
Another research activity involves the lab of Associate Professor Mohammad Jahanshahi at Purdue University. With the help of other researchers, Associate Professor Jahanshahi and Associate Professor Cruz published the results of their first collaborative study in 2021, a special fusion of agriculture and engineering focused on another high-profile fungal disease called wheat blast. The model developed detected and classified images of disease symptoms within two seconds.
Jahanshahi says they are working together to train the next generation of problem solvers: “Whether they're working on civil infrastructure health monitoring or on the epidemiology of agricultural diseases, they can have a big impact on not only humans but the environment.”
For more information, visit ag.purdue.edu.
Purdue University's College of Agriculture is one of the world's leading universities in agricultural, food, life and natural resource sciences. Read all of the author's stories here.
