Classic microscopes give a modern twist. US researchers are developing AI-powered microscope systems that allow farmers and land managers around the world to make soil health testing faster, cheaper and more accessible.
Researchers at the University of Texas at San Antonio, USA, have been able to combine low-cost optical microscopes with machine learning to measure the presence and amount of fungi in soil samples. Their early stage proof-of-concept technology will be presented at the Goldschmidt conference in Prague on Wednesday, July 9th.
Determining soil fungi's abundance and diversity plays an important role in nutrient biogeochemical cycling, moisture retention and plant growth, so the richness and diversity of soil fungi can provide valuable insight into soil health and fertility. This knowledge allows farmers to optimize crop production and sustainability by making informed decisions about soil management, such as fertilizer application, irrigation, and cultivation.
Optical microscopes are the oldest design of microscopes and have been used for a long time to discover and identify small organisms in the soil. Other forms of soil testing use techniques such as phospholipid fatty acid testing and DNA analysis to detect organisms, or measure the presence of chemicals such as nitrogen, phosphorus, and potassium. Although powerful, these modern methods tend to be costly or simply emphasize chemical composition, often overlooking the complete biological complexity of soil ecosystems.
Alec Graves of the University of Texas San Antonio University University of Science presents his research this week at the Goldschmidt conference. He said: “Current forms of biological soil analysis are limited and require either expensive laboratory equipment to measure molecular composition or experts who identify organisms per field of view using laboratory microscopes. Comprehensive soil testing is not widely accessible to farmers and land managers who need to understand how agricultural practices affect soil health.
“We are using machine learning algorithms and optical microscopes to create low-cost solutions for soil testing that reduces the required labor and expertise while providing a more complete picture of soil biology.”
In the early stage design, researchers built and tested machine learning algorithms to detect fungal biomass in soil samples and incorporated them into custom software to label microscopic images. This was created using a dataset of images of thousands of fungi from soils in southern Texas. The software works with just 100x and 400x total microscope magnifications. This is available in many affordable off-the-shelf microscopes, including those found in school labs.
“Our technology analyzes videos of soil samples, analyzes them into images, and uses neural networks to identify and quantify fungi,” Graves says. “Our proof of concept can already detect fungal chains in diluted samples and estimate fungal biomass.”
The team is currently working to integrate the technology into a mobile robotics platform for detecting fungi in the soil. The system combines sample collection, microportography and analysis into a single device. They aim to be ready to test fully developed deployable devices within the next two years.
The study is led by Professor Saugata Datta, Director of Sustainability and Policy at UTSA Water Institute, and is funded by the USDA National Resource Conservation Service. Details about the machine learning algorithm will be published in a peer-reviewed journal later this year.
The Goldschmidt Conference is the world's Geochemistry Conference. This is a joint council of the European Geochemical Society and the Geochemical Society (USA), with over 4,000 representatives participating. It will be held in Prague, Czech Republic from July 6th to 11th, 2025.
/Public release. This material of the Organization of Origin/Author is a point-in-time nature and may be edited for clarity, style and length. Mirage.news does not take any institutional position or aspect, and all views, positions and conclusions expressed here are the views of the authors alone.
