“Insect Taxonomy 4.0: Lenses of Multiple Realities (Integration of Diagnostic Imaging Techniques, Computational Models, and Molecular Validation in the Era of Artificial Intelligence),” presented by PhD student Mohammed Maytham Abdulhai, under the supervision of Assist. Prof. Mohsen Abdul Ali Mohsen and Assist. Prof. Istabraq Mohammed Abdul Ridha. This seminar was held as part of the department’s scientific activities to keep pace with modern technological developments in the field of plant protection.
The seminar addressed a key issue related to the challenges facing traditional taxonomy, where the researcher indicated that approximately 80% of insects remain scientifically undescribed.
He highlighted the risk of complete morphological similarity that can mislead the naked eye, resulting in misdiagnosis and the application of ineffective pesticides, which in turn leads to significant economic and environmental losses.
The seminar also reviewed the limitations of classical morphology, which relies heavily on human judgment and faces difficulties in distinguishing between geographically similar species. To overcome these challenges, the concept of a shift toward “Integrative Taxonomy” was introduced, which combines three main dimensions:
The physical world (diagnostic imaging): the use of advanced techniques such as scanning electron microscopy (SEM) to study fine structures, micro-computed tomography (Micro-CT) to build three-dimensional models of internal organs without destroying the sample, and hyperspectral imaging for early detection of plant stress.
The digital world (artificial intelligence): the application of computer vision and deep learning techniques for real-time pest identification in the field through “attention mechanisms” that focus on distinguishing features such as wings while ignoring background noise.
The molecular world (DNA barcoding): the use of DNA barcoding as a gold standard for resolving the identity of cryptic species and overcoming limitations across developmental stages (egg, larva, adult), with particular emphasis on the mitochondrial COI gene.
The seminar also discussed the “cyber workflow,” which represents the integrative laboratory loop, starting from digitization, followed by automated analysis where algorithms propose a preliminary classification, proceeding to molecular extraction in cases of anomalies, and concluding with cyber documentation by uploading the “digital twin” and genetic fingerprint to global biodiversity cloud platforms.
In conclusion, it was emphasized that artificial intelligence serves as a preliminary screening tool that accelerates the process hundreds of times, but it does not replace expert judgment or DNA validation. This integration supports sustainable food security by reducing pesticide use by up to 30–50% through precise diagnosis and intelligent monitoring.
