In the heart of Burkina Faso, a groundbreaking study is revolutionizing the way cassava viral diseases are diagnosed and managed, with significant implications for food security and agricultural sustainability. Led by Seydou Sawadogo from the Laboratory of Plant Virology and Biotechnology at the National Institute of Environment and Agricultural Research (INERA) in Ouagadougou, this research leverages artificial intelligence and participatory surveillance to combat cassava mosaic disease (CMD) and cassava brown streak disease (CBSD), which have long plagued Sub-Saharan Africa’s cassava crops.
The study, published in *Frontiers in Plant Sustainability Systems* (translated from French), introduces a low-cost, early detection method for cassava viral diseases using the Plantvillage Nuru app, an AI-based diagnostic tool. This innovative approach involves farmers, agricultural extension agents (AEA), and disease diagnosis experts, creating a collaborative network that enhances disease surveillance and management.
“Participatory approaches are crucial in plant health,” Sawadogo explains. “They empower local communities and agricultural extension agents to take an active role in disease surveillance, leading to early detection and more effective management strategies.”
The research involved training 60 AEAs to identify CMD, CBSD, and cassava green mite (CGM) symptoms using the AI tool. Equipped with smartphones, these agents conducted field surveillance, either through visual inspection or with the AI tool. The participation rate of the AEAs surged to 60%, and the number of surveyed fields increased to 132, demonstrating the effectiveness of the training and smartphone allocation.
The results were promising. The AI tool’s diagnostic accuracy was comparable to that of visual inspections, with no significant difference in CMD detection rates. The mean scores for CMD detection were 29.83% for the AI tool, 37.12% for experts, and 36.10% for molecular analysis among AI tool users. For visual inspection users, the mean scores were 46.07% for AEAs and experts’ perception and 43.87% for molecular analysis.
While the AI tool misdiagnosed 5% of CMD cases as CBSD, molecular analysis confirmed these as CMD, highlighting the tool’s potential for improvement. The study found that 31.06% of the fields were infected with CMD, with the African Cassava Mosaic Virus being the predominant strain (93.33%).
This research underscores the potential of participatory surveillance and AI tools in early disease management, offering a scalable and cost-effective solution for plant health monitoring. As Sawadogo notes, “The integration of AI and participatory approaches can transform how we manage plant diseases, ensuring food security and sustainability.”
The implications for the agricultural sector are profound. By empowering local communities and leveraging technology, this approach can enhance disease surveillance, reduce yield losses, and improve food security. The study’s findings pave the way for future developments in plant health management, highlighting the importance of collaborative efforts and technological innovation in addressing agricultural challenges.
As the world grapples with the impacts of climate change and increasing food demand, innovative solutions like this one are crucial. The research published in *Frontiers in Sustainable Food Systems* not only advances our understanding of cassava viral diseases but also sets a precedent for future agricultural technologies, shaping the future of sustainable food systems.