Assessment of Peripheral Blood Cell Composition in Mice with Traumatic Brain Injury Using Artificial Intelligence Technologies
Year: 2026
Type: article
Abstract: Introduction. Traumatic brain injury (TBI) is one of the main causes of mortality and disablement in humans and animals worldwide. Early assessment of TBI severity enables reducing deferred adverse effects and mortality. However, implementation of artificial intelligence (AI) technologies in clinical practice is hindered due to the shortage of qualified personnel and lack of specialized datasets in blood cytology. The aim of the study is to analyse changes in cellular composition of peripheral blood in mice after severe TBI and to create a dataset of blood smears to be used in AI algorithms. Materials and М ethods. The study was conducted at Don State Technical University in the period from November to December 2025. The objects for the study were images of the peripheral blood smears obtained from mice with TBI: blood samples from the caudal vein were collected, and smears were prepared and digitized. Image labeling was performed using the PULLAR platform (a development of “MedTsifra” Research Laboratory, Rostov-on-Don). Results. A dataset of TBI blood smears was created, and 60 images containing neutrophils, monocytes, and lymphocytes were labeled. Elevated counts of segmented neutrophils and monocytes indicate the development of the acute inflammatory response in traumatic brain injury condition. A decrease in lymphocyte count is consistent with the data about reduction of the lymphocyte numbers at the early TBI stages and indicates the state of immunosuppression. Discussion and Conclusion. The obtained data demonstrate the informative value of the peripheral blood cell composition analysis in mice with traumatic brain injury, which confirms the potential of developing and implementing AI technologies for improving the quality of diagnostics and automating the routine tasks.
Authors: M. S. Kaplya, Y. P. Gordeeva, S. V. Rodkin
Institution: Don State Technical University
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W7225880751