Stride by stride, breath by breath: Sensor-based assessment of locomotion and respiration in harness racehorses to optimize welfare and performance

Abstract

Horses are widely used for sport and recreational activities, placing high physical demands on their bodies. During exercise, multiple body systems—including the musculoskeletal, respiratory, and cardiovascular systems—continuously adapt and function as an integrated whole. Dysfunction in any one of these systems can negatively affect both performance and welfare, highlighting the importance of objective and integrated monitoring of horses during exercise. Although technological advances have made it possible to measure many physiological parameters in horses, a practical and coherent picture of the horse in motion is often still lacking. This thesis aims to bridge that gap by identifying and objectively evaluating non-invasive technologies for real-time monitoring of gait and respiration during exercise. This research shows that reliable technologies for monitoring heart rate and locomotion are already widely available, whereas practical tools for assessing respiratory function remain limited. By using wearable sensors, such as inertial measurement units (IMUs), and microphones, subtle exercise-induced changes in both movement and breathing can be detected. The results demonstrate that trotters use different locomotor strategies at high speeds and that intense exercise leads to measurable changes in movement patterns and stride variability. A key innovative aspect of this work is the use of microphones combined with artificial intelligence to automatically analyse respiration sounds. This approach enables reliable, field-based monitoring of respiratory patterns without burdening the horse. Overall, this thesis contributes to improved monitoring strategies, deeper insight into equine exercise responses, and a better balance between performance and welfare, with the horse’s perspective in mind.