The article presents a broad analysis of methods for applying artificial intelligence to predict injuries and fatalities in the Thoroughbred industry. The study is based on an interdisciplinary approach that incorporates elements of veterinary medicine, sensor technologies, radiomics, image analysis, and the processing of clinical records. Particular attention is given to a comparative-analytical review of modern models, including computer vision techniques, radiomic analysis (μCT), machine learning algorithms, and deep neural network architectures (CNN, ANN), as well as the use of inertial measurement units (IMU) for quantitative risk stratification. The limitations of individual approaches are identified: small sample sizes and low classification stability for eye regions in computer vision systems, high cost and limited accessibility of radiomics, and the dependence of sensor technologies on track surface and conditions. The integration of domain transfer methods, facial expression analysis, and NLP services for unstructured medical data processing is emphasized. The impact of multimodal integration of images, sensor metrics, and clinical records on improving the accuracy and robustness of predictions is examined. The need to shift from universal monitoring strategies to targeted control of high-risk groups is substantiated, enabling proactive equine health management and reducing fatality rates in the racing industry. The article proposes an original classification of promising development directions, including complex architectures, the use of Retrieval Augmented Generation to link sensor and clinical data, and the implementation of integrated platforms in the practice of insurance, breeding, and injury prevention. The material will be of interest to researchers in veterinary medicine, sports analytics, artificial intelligence, and digital technologies, as well as to specialists engaged in the development and implementation of risk prediction systems in the equine industry.

Artificial intelligence, equine industry, injury prediction, fatal musculoskeletal injuries, computer vision, inertial measurement units, multimodal data integration

Anderson, K., Morrice-West, A. V., Walmsley, E. A., Fisher, A. D., Whitton, R. C., & Hitchens, P. L. (2023). Validation of inertial measurement units to detect and predict horse behaviour while stabled. Equine Veterinary Journal, 55(6), P. 1128–1138. https://doi.org/10.1111/evj.13909

Asti, V., Ablondi, M., Molle, A., Zanotti, A., Vasini, M., & Sabbioni, A. (2024). Inertial measurement unit technology for gait detection: A comprehensive evaluation of gait traits in two Italian horse breeds. Frontiers in Veterinary Science, 11, Article 1459553. https://doi.org/10.3389/fvets.2024.1459553

Basran, P. S., McDonough, S., Palmer, S., & Reesink, H. L. (2022). Radiomics modeling of catastrophic proximal sesamoid bone fractures in Thoroughbred racehorses using μCT. Animals, 12(21), 3033. https://doi.org/10.3390/ani12213033

Bogossian, P. M., Nattala, U., Wong, A. S. M., & others. (2024). A machine learning approach to identify stride characteristics predictive of musculoskeletal injury, enforced rest and retirement in Thoroughbred racehorses. Scientific Reports, 14, Article 28967. https://doi.org/10.1038/s41598-024-79071-1

Broomé, S., Ask, K., Rashid-Engström, M., Haubro Andersen, P., & Kjellström, H. (2022). Sharing pain: Using pain domain transfer for video recognition of low grade orthopedic pain in horses. PLOS ONE, 17(3), e0263854. https://doi.org/10.1371/journal.pone.0263854

Crecan, C. M., & Peștean, C. P. (2023). Inertial sensor technologies—their role in equine gait analysis: A review. Sensors, 23(14), 6301. https://doi.org/10.3390/s23146301

Feighelstein, M., Riccie-Bonot, C., Hasan, H., Weinberg, H., Rettig, T., Segal, M., Distelfeld, T., Shimshoni, I., Mills, D. S., & Zamansky, A. (2024). Automated recognition of emotional states of horses from facial expressions. PLOS ONE, 19(7), e0302893. https://doi.org/10.1371/journal.pone.0302893

Kim, S. M., & Cho, G. J. (2023). Analysis of various facial expressions of horses as a welfare indicator using deep learning. Veterinary Sciences, 10(4), 283. https://doi.org/10.3390/vetsci10040283

Lencioni, G. C., de Sousa, R. V., de Souza Sardinha, E. J., Corrêa, R. R., & Zanella, A. J. (2021). Pain assessment in horses using automatic facial expression recognition through deep learning-based modeling. PLOS ONE, 16(10), e0258672. https://doi.org/10.1371/journal.pone.0258672

Mc Sweeney, D., Wang, Y., Palmer, S. E., Holmströem, M., Donohue, K. D., Farnsworth, K. D., Sanz, M. G., Lambert, D. H., & Bayly, W. M. (2025). Thoroughbreds deemed to be most at risk by inertial measurement unit sensors suffered a fatal musculoskeletal injury at a higher rate than other racehorses. Journal of the American Veterinary Medical Association. Advance online publication. https://doi.org/10.2460/javma.25.04.0268