Identifying lameness in horses through deep learning.

Wang Y, Li J, Zhang Y, Sinnott RO. Identifying lameness in horses through deep learning. SAC '21: Proceedings of the 36th Annual ACM Symposium on Applied Computing. 2021; Pages 976-985. https://doi.org/10.1145/3412841.3441973

Summary: This study investigates the use of deep learning for markerless motion tracking to detect lameness in horses. A neural network was trained to identify specific body parts of the horse, allowing for analysis of motion trajectories to assess lameness. Two state-of-the-art convolutional network models were evaluated for their effectiveness in animal pose estimation. The study also developed a mobile application to test the feasibility of using this technology in real-world scenarios, offering a time-efficient alternative to traditional methods that require trackable markers and high-speed cameras.