Wageningen University & Research in the Netherlands has achieved a significant breakthrough in the automated identification of plant-parasitic nematodes, a development poised to aid global agriculture. Researchers have developed an artificial intelligence system capable of identifying the root-knot nematode Meloidogyne chitwoodi with 96% accuracy, matching the performance of an experienced taxonomic nematologist.
The achievement addresses a critical bottleneck in plant health management. Accurate nematode species identification is complex, costly, and requires specialized expertise, limiting diagnostic capacity worldwide. Harmful nematodes, such as stem nematodes and root-knot nematodes, can render crops like ornamental flower bulbs, onions, and seed potatoes unsalable or unexportable, causing billions of euros in damages annually, with estimates suggesting 10% of global agricultural output is affected.
The AI system’s initial focus on Meloidogyne chitwoodi was deliberate, according to researcher Leendert Molendijk. “If it works for a species as difficult as this one, it should also work for species that are easier to distinguish,” he stated in a press release from Wageningen University & Research.
The project, conducted in collaboration with agrotech company Veridi Technologies BV, relied on a substantial collection and validation of microscopic images. Researcher Pella Brinkman emphasized the importance of reliable biological data in training the AI model. “We provided nematodes from our cultures and identified them accurately. What we have is essential for training the AI model on a large number of images. We also performed validation checks to assess reliability,” Brinkman explained.
The process of identifying errors made by the AI system also refined the morphological criteria used by the algorithm, improving its overall accuracy. This development is particularly significant for agricultural regions lacking extensive taxonomic expertise, such as Morocco, where Meloidogyne species pose a major threat to vegetable crops, citrus fruits, and greenhouse production.
Researchers believe the technology has the potential to transform global plant health surveillance. “If People can apply it to other nematode species, it could have a significant impact worldwide,” Molendijk said. The automation of nematode identification could strengthen diagnostic capabilities, improve understanding of soil health, and support more targeted integrated pest management strategies.
Ongoing research at Wageningen University & Research, detailed in a 2020 publication in Frontiers in Plant Science, explores the complex relationship between plant roots and the surrounding microbial community, including nematodes. This work highlights the importance of understanding the active rhizobiome – the community of microorganisms directly interacting with plant roots – and how it shifts in response to different soil management practices.
