New paper entitled “Using deep learning to assess the toxicological effects of sublethal exposure of a novel green pesticide in a stored-product beetle
New paper entitled “Using deep learning to assess the toxicological effects of sublethal exposure of a novel green pesticide in a stored-product beetle”, 2026, by Anita Casadei, Maria C. Boukouvala, Gianluca Manduca, Nickolas G. Kavallieratos, Filippo Maggi, Marta Ferrati, Eleonora Spinozzi, Cesare Stefanini, Antonio DeSimonea, and Donato Romano
at Pest Management Science, Volume 82, pages 4319 – 4331, DOI 10.1002/PS.70545, wileyonlinelibrary.com
Τhe paper “Using deep learning to assess the toxicological effects of sublethal exposure of a novel green pesticide in a stored-product beetle”, 2026, by Anita Casadei, Maria C. Boukouvala, Gianluca Manduca, Nickolas G. Kavallieratos, Filippo Maggi, Marta Ferrati, Eleonora Spinozzi, Cesare Stefanini, Antonio DeSimonea, and Donato Romano (https://doi.org/10.1002/ps.70545) has been selected as a cover page of Issue 7 of the Volume 82 (https://doi.org/10.1002/ps.70990) of the international journal Pest Management Science (Impact Factor = 3.8, Q1 in Entomology, 7/110). It should be noted that the Editors of Pest Management Science selected the aforementioned paper among the 97 papers that have been published in the Issue of July 2026.
This study analyses the sublethal effects of the natural substance carlina oxide on Prostephanus truncatus (Coleoptera: Bostrychidae), providing new insights into its behavior through a multidisciplinary approach. For this purpose, a fully automated computer vision method was developed for the digital tagging of two specific parts of the insect's body, enabling the generation of an annotated dataset without manual intervention. This data was used to train a Convolutional Neural Network (CNN) for pose estimation. A second, specialized CNN focused on the antennae to investigate neurosensory variations. Motor analysis showed that the LC30 concentration of carlina oxide reduced average speed and distance, altered exploratory behavior, and affected thigmotaxis. Statistically significant features were evaluated using machine learning models Random Forest, SVM, and KNN, distinguishing control groups from groups under the influence of LC30 and LC10. The findings indicate the possibility to apply the current method to green or/and synthetic pesticides, obtaining a wider evaluation about their toxicological effects on the motor behavior of different insect species.
The Greek research team is framed by Dr Maria C. Boukouvala, PostDoc Researcher of the Laboratory of Agricultural Zoology and Entomology, Agricultural University. The research was carried out in the context of Dr Boukouvala's participation in the Short-Term Scientific Mission (STSM) of the European Program - Network COST CA20132 - Urban Tree Guard - Safeguarding European urban trees and forests through improved biosecurity, (STSM Request Reference No. E-COSTGRANT-CA20132-f2fdf865). The experimentation was carried out at the Institute of BioRobotics, Sant'Anna School of Advanced Studies (Pontedera, Italy) from 1st to 30th September 2024, under the supervision of Prof. Donato Romano. In addition, this research was supported by the following scientific programs: HORIZON-EIC-2023-PATHFINDEROPEN-01, SENSORBEES ‘Sensorbees are ENhanced Self-ORganizing Bio-hybrids for Ecological and Environmental Surveillance’; the Italian Space Agency (ASI) DC-DSR-UVS-2022-375 Project ‘pRomoting pEdogenesis throuGh lunar sOil-terrestriaL organIsms interaction For moon FErtilization - REGOLIFE’ [ASI N.: 2024-7-U.0; CUP: J83C24000310005]; and the BRIEF “Biorobotics Research and Innovation Engineering Facilities” (identification code: IR0000036), a project funded under the National Recovery and Resilience Plan (NRRP), Mission 4 Component 2 Investment 3.1 of the Italian Ministry of Universities and Research, with funding from the European Union - NextGenerationEU.
Scientists from five academic and research institutions in Greece, Italy and the United Arab Emirates participated in the research paper: The BioRobotics Institute, Sant'Anna School of Advanced Studies, Pisa, Italy, Department of Excellence in Robotics and AI, Sant'Anna School of Advanced Studies, Pisa, Italy, Laboratory of Agricultural Zoology and Entomology, Department of Crop Science, Agricultural University of Athens, Attica, Greece, ChIP-Chemistry Interdisciplinary Project Research Center, School of Pharmacy, University of Camerino, Camerino, Italy, Department of Robotics, Mohamed bin Zayed University of Artificial Intelligence, Masdar, United Arab Emirates.
