Machine learning in assessing canine bone fracture risk: A retrospective and predictive approach

Kostenko, E., Šengaut, J., & Maknickas, A. (2024). Machine learning in assessing canine bone fracture risk: A retrospective and predictive approach. Applied Sciences, 14(11), 4867. https://doi.org/10.3390/app14114867

This study addresses the challenge of predicting bone fracture risk in canines, particularly in extra-small breeds and young dogs. A machine learning algorithm using a random forest classifier was developed and trained on 2,261 cases, incorporating factors such as age, gender, breed, and weight. The model successfully predicted fracture risk, although the results are considered preliminary due to the dataset's size. The authors highlight the tool's potential to improve veterinary care by enabling early detection and personalized preventive measures for high-risk patients.