CETIM contributes its expertise to the project by recreating the process of infection of the plant by the fungus and developing disruptive technologies, based on Artificial Vision and Artificial Intelligence, that allow the early detection of mycotoxins.

Mycotoxins are a problem for human and animal health. These toxic substances are produced by a wide variety of fungi and affect different foodstuffs such as spices, cereals, fruits and nuts, meat, dairy products, etc. The accumulation of these mycotoxins in the body can lead to chronic effects including the development of tumours.

Currently, the identification of the presence of fungi in raw materials is done visually at the entry of the material into the plant itself and after storage prior to the processing of the product. Therefore, the identification of a possible affectation depends directly on the experience and judgement of the operators. In addition, if there is no early detection and elimination measures, the consequences are significant losses and high economic costs in the agricultural industry.

In order to find solutions to this problem, Micovision is born, a project in which CETIM is working in collaboration with GENERANDI, to find solutions that allow a safe and early detection of mycotoxins. To this end, an artificial vision system is being designed to detect fungal damage to aromatic and medicinal plants intended for food use.

The main objective of the project is to improve the productivity of a seasoning and herbal tea production plant, avoiding losses of raw material due to damage caused by fungal growth, as well as to improve food safety throughout the product marketing chain by avoiding the presence of mycotoxins produced by fungi.

CETIM Expertise

At CETIM we provide our expertise in mycovision. Firstly, by evaluating two technologies that allow us to establish an automatic detection system: artificial vision techniques and the use of spectroscopic measurements.

Secondly, we will infect two aromatic plants (oregano and basil) with two different species of fungi in order to take images showing the evolution of the disease. This image capture and labelling, which will be carried out in our Technology Centre, will allow the development of a dataset for the training of an Artificial Intelligence algorithm, which will be carried out by GENERANDI, leader of the project.

Likewise, we will provide support in the validation phase of the technologies developed and we will carry out a bibliographic study on the mechanisms of fungal contagion and propagation, as well as existing methodologies for their detection in crops.

In short, Micovision, which has been financed by the Ministry of Industry, Trade and Tourism through the European Union’s Recovery and Resilience Mechanism and NextGenerationEU funds, will enable the development of an automated diagnostic system based on Artificial Intelligence for the detection of fungi in digital matter. Thus, it will not only be the experience and judgement of the operators who will decide the existence of mycotoxins, but the technologies developed will allow an early detection of the presence of this type of toxic substances in plants. All this, increasing cost savings and avoiding large losses in the agri-food industry.