AI-Driven Deep Learning Models for Detecting Anomalies in Radiographic Images of Sport Horses

Details

Supervisors

Vanderdonckt, Jean

Faculty

Louvain School of Management

Degree label

Master [120] : ingénieur de gestion, à finalité spécialisée

Abstract

The equestrian industry is a major global market with substantial economic impact, significantly contributing to various national economies. In Europe alone, the industry generates approximately €112 billion, highlighting its importance within the global horse market. A key aspect of this sector is horse trading, where the financial stakes can be exceedingly high. Foals and competition horses often fetch substantial sums, underscoring the importance of accurate pre-purchase assessments. Currently, these assessments rely heavily on subjective veterinary opinions, which traditionally do not utilize comprehensive scientific data. This thesis addresses the gap in technological advancement and objectivity within veterinary evaluations by introducing a data-driven approach to horse health assessment. Leveraging machine learning techniques, this research proposes a novel method for analyzing and predicting the health and performance of horses, thereby reducing human subjectivity in critical transactions. By incorporating mathematical and scientific principles into the evaluation process, this thesis aims to revolutionize the equestrian sec- tor, offering a more objective and reliable framework for horse trading and performance assessment. This advancement has the potential to significantly impact decision-making processes and improve the accuracy of health assessments in the industry.