PG-data

This study introduces a novel approach for the early detection of induced colic in horses through the use of accelerometer devices. By inducing transient colic in experimental mares and collecting accelerometric data, we developed a multi-phase strategy to identify episodes of colic and categorize their severity. The model utilizes both behavioral and activity index characteristics to detect the presence of colic and assess its severity. Our results demonstrate high accuracy in distinguishing between normal and pain-related behaviors, offering significant potential for non-invasive, continuous monitoring of horse health, and potentially faster, more effective treatment of colic. The main limitation is the inclusion of a limited number of horses with severe pain-related behaviors in the dataset.