AI-powered platform improves diagnosis of Asian soybean rust.

Brazilian scientists have developed a platform To diagnose Asian soybean rust, one of the most severe diseases affecting the crop, the technology integrates artificial intelligence (AI) with the combined analysis of climatic, agronomic, and digital image data. The cloud-based system assesses the risk of disease occurrence and generates reports with technical management recommendations, contributing to more precise decisions in the field.

The tool collects data from environmental sensors, digital leaf images, and agronomic parameters such as cultivar, spacing, and sowing time. The results are presented on an online dashboard, allowing farmers to track time series of climate data and plant images.

The system was developed as part of the project. Advanced Digital Tool for Agricultural Risk Management , with funding from the São Paulo State Research Foundation ( Fapesp The initiative was part of the doctoral research of computer scientist Ricardo Alexandre Neves at the Federal University of São Carlos ( UFSCar ), with Paulo Cruvinel , researcher at Embrapa Instrumentation (SP), as advisor. 

The study "A cloud-based intelligence system for risk analysis of Asian soybean rust in soybean crops" It was published in July 2025 by the magazine. AgriEngineering . 

Data fusion facilitates diagnosis.

Scientists developed the system through research on farms, using a model that incorporates climatic variables, data related to soybean plants, and data obtained from digital images of soybean leaves. The climatic variables were observed during the area's monitoring period.

“The technology classifies the favorability of the disease into three levels — low, medium, and high — depending on the combination of variables related to the stage of infestation. This allows for more effective and precise diagnoses and prognoses for disease control,” adds Neves. According to him, the favorability level is defined by statistical inference based on the behavior of the set of variables. 

The researchers explain that the system works by combining data. The main data allows for the analysis of factors essential to the development of the fungus, such as the period of leaf moisture — relative humidity above 90%, in the temperature range between 15°C and 28°C — or the dew point. 

This work utilizes advanced and specific processing techniques to extract information from digital images of soybean leaves. Color patterns, such as green, yellow, and brown, are associated with the stages of disease progression.

Cruvinel reports that, to integrate the data, the study evaluated two methods. In the end, the choice for the system fell on the Hidden Markov Chain model, which provides robustness, effectiveness, and efficiency to the decision-making process. This methodology proved superior to fuzzy logic, achieving 100% accuracy in matching the evaluated scenarios for the risk of Asian soybean rust occurrence in soybean cultivation areas. 

"The model developed to combine data from different variables made it possible to structure a complete set of rules that systematically considers different situations in which the disease is likely to occur," says the researcher.

During the four-year study with the conventional cultivar BRS 536 soybean variety from Embrapa The researchers used more than 2 gigabytes of data per crop cycle, considering information collected in real fields during cultivation, in georeferenced plots in the Poxoréu-MT region and photographed under known light levels. 

Data available to farmers on the web.

The analytical reports available on the control panel were compiled based on twenty years of historical data and allow for the evaluation of crop cycle periods. The system has a user-friendly interface, organized with basic information of interest to farmers and potential users.  

According to Cruvinel and Neves, the reports aim to support farmers' decision-making regarding the management of cultivated areas, allowing them to assess the occurrence or absence of Asian soybean rust and the severity of the disease. Furthermore, they offer agronomic recommendations based on the diagnosis for disease control.

Cruvinel adds that the reports can be found in the “Agricultural Recommendations” tab of the control panel, where there is also a link to the website. AGROFIT, a database with information on agrochemicals and related products that have been registered by the Ministry of Agriculture and Livestock ( Map ) from Brazil, for consultation and selection of fungicides recommended for the control of Asian soybean rust. 

The solution reduces the use of fungicides.

Researchers say the system makes it possible to monitor the presence or absence of Asian soybean rust, as well as assess the dynamics of the disease's occurrence at different stages of severity and risk in the agricultural production process.

“The key point of the research was to create a method that integrates heterogeneous data to provide a more reliable diagnosis. Relying solely on images or isolated climate data is not sufficient for an accurate assessment, which can lead to false-positive diagnoses. Furthermore, the solution offers prevention and rational use of fungicides,” states Neves, who is currently a professor at the Federal Institute of São Paulo ( IFSP ), campus São João da Boa Vista.