The poultry industry is pivotal in meeting the food demands of a growing global population. However, the welfare of the chickens is a critical concern, as poor living conditions can lead to abnormal behaviors that adversely affect the health and productivity of the entire flock. To maintain and enhance the chickens’ health, it is essential to implement an automated surveillance system that monitors the behaviors of the chickens in poultry farms. This paper emphasizes the critical need for a surveillance system in poultry farming that leverages the capabilities of the pre-trained YOLOv8 pose model, in order to identify the chickens and capture the necessary key-points. The novel system proposed captures the key-points of chickens, aiding in the detection and classification of wryneck disease, a common condition in the poultry industry. The implementation of such advanced technology promises significant improvements in the poultry sector by enabling farmers to swiftly intervene, minimizing morbidity and mortality rates. Furthermore, the system’s utilization of YOLOv8 for pose estimation offers advantages in terms of speed and accuracy, enhancing the surveillance efficiency of poultry farms. The insights provided in this study underscore the potential of applying computer vision machine learning techniques to animal health monitoring, contributing to the sustainability of the poultry industry and the well-being of the birds. The proposed system produces an accuracy of 99.72% for frame by frame classification of chickens with wry neck disease, and healthy chickens.