Diagnostic error in veterinary imaging: an underexplored field and future directions for observer performance research

Diagnostic error is increasingly recognized as an important component of diagnostic quality in human radiology, where decades of research have established observer-performance as a measurable scientific process shaped by perceptual, cognitive, organizational, cultural, and technological factors. In veterinary diagnostic imaging, however, diagnostic error remains comparatively under-investigated, and existing evidence is fragmented across studies of diagnostic accuracy, observer variability, image quality, and quality assurance. In this perspective, we examine how concepts developed in human radiology may inform the study of diagnostic error in veterinary imaging, while considering the distinctive clinical, methodological, organizational, and ethical characteristics of veterinary practice. We argue that veterinary diagnostic imaging would benefit from recognizing diagnostic error as a distinct field of scientific investigation and from developing observer-performance research as a systematic area of investigation. We identify four interconnected priorities: developing robust methods to measure diagnostic performance, investigating the mechanisms underlying diagnostic error, evaluating interventions to improve observer performance, and building collaborative and multidisciplinary research networks. Together, these priorities provide a basis for moving from the recognition of diagnostic discrepancies toward a systematic science of diagnostic error aimed at improving diagnostic performance, quality assurance, education, and ultimately patient care and welfare.