Monitoring integrity of hazelnuts by Terahertz Imaging

The ability of Terahertz Imaging to visualize the water content in small objects makes it the ideal technique to non invasively monitor the health status of plants by analysing their leaves, but also to assess the quality of dry fruits based on the identification of areas where the water amount is changed due to e.g. fungi infections, insect bites or other causes.

Dr. Mario Pagano visited the Department of Physics at the University of Pisa, one of the facilities of our MultiModal Molecular Imaging Italian Node (MMMI), to apply the technique for the quality assessment of hazelnuts, and more specifically to investigate the potential to discriminate between healthy and “cimiciato” hazelnuts.

Dr. Mario Pagano is a permanent researcher at the Research Institute on Terrestrial Ecosystems (IRET) of the National Research Council of Italy (CNR), based in Florence (Italy). His research focuses on plant physiology, including the assessment of plant responsiveness to abiotic stress and other factors such as for example exposure to sound waves or ozone.

The imaging instrumentation and the technical and scientific skills of the staff at the MMMI Node in Pisa are the perfect match to his own competencies and research interests. This gave a start to a long-lasting collaboration resulting in some interesting scientific papers demonstrating how Terahertz Imaging can be used in plant imaging and have potential interesting applications for food quality assessment.

When Terahertz imaging was made available in open access through Euro-BioImaging, Dr. Pagano applied for access with a project on the feasibility of using the technology for assessing the quality of post harvested hazelnuts.

Before complete ripening, hazelnuts are often bitten by an insect, the brown marmorated stink bug (Halyomorpha halys). This insect inoculates hydrolizing enzymes into the fruits, that are able to do a sort of pre-digestion of the fruit matrix, allowing successive easy sucking by the bug. This causes a damage in the fruit, which will then present a high incidence of the characteristic off-flavour defined as “cimiciato”. This damages hazelnut cultivations and has a negative economic impact. The hazelnut manufacturers perform quality checks on the fruits, but currently this can be done only manually, even though the damage is not always detectable by visual inspection, or using very expensive equipment.

Since during the pre-digestion process the water content in the hazelnut changes, Dr. Pagano had the idea to investigate if Terahertz Imaging would be suitable to distinguish between healthy and “cimiciato” hazelnuts. Indeed, the Terahertz Imaging equipment is quite cheap, easily movable, and could be quickly included in the industrial chain.

Together with Prof. Alessandra Toncelli at the University of Pisa, Dr. Pagano performed Terahertz imaging acquisitions on both healthy and “cimiciato” hazelnuts on a Terasense instrument, and found that below a given value of the signal attenuation in the images it is indeed possible to distinguish the two groups. Based on the acquired data, Dr. Pagano and colleagues developed a predictive model which is capable of correctly identifying 100% of “cimiciato” fruits in unknown samples, showing a perfect specificity. Currently, some of the healthy ones are also identified as “cimiciato”, thus the technique should be further optimized to avoid wasting healthy hazelnuts in the productive process. Nevertheless, the results are very promising and suggest that Terahertz Imaging can indeed be a viable approach for the recognition of defective hazelnuts before their commercialization or processing, and since this method is experimentally simple, automatable, it requires a low-cost apparatus and can potentially be implemented in real-time, it can be of great interest for the food industry.

You can read more on this work here: https://doi.org/10.1016/j.heliyon.2023.e19891.

Dr. Pagano is currently working to develop an automated, machine learning based approach to improve and fasten the predictive process. Dr. Pagano is very happy with the support he received from the MMMI Node: “Having access to the Terahertz Imaging facility in Pisa was crucial for my research on applying Terahertz radiation to plants and dry fruits, as it provided the necessary equipment and technical support for this type of study”, he says. So happy that he recently applied for access again, this time with a project on abiotic stress in plants, monitored through Terahertz imaging of the leaves, and first results are again quite promising. Stay tuned to learn more about this project when it is completed!