Quo vadis ultrasound: toward a terminologia sonographica.

The increasing use of artificial intelligence, automated image analysis, and robotic systems in ultrasound diagnostics necessitates a consistent, physically grounded, and cross-device comparable terminology. However, contemporary ultrasonographic findings are frequently described using historically evolved, metaphorical, and ambiguously defined terms. This semantic variability limits reproducibility, data harmonization, and the learning and reporting capacity of AI-based models. Ultrasound images do not represent direct anatomical depictions, but rather algorithmic reconstructions of acoustic interactions based on model-dependent assumptions. Artifacts and echo patterns, therefore, reflect the discrepancy between physical reality and mathematical approximation within signal processing. Inconsistent nomenclature widens the gap between measured variables, reconstruction models, and clinical interpretation. This paper analyses this structural problem and outlines a development framework for an internationally compatible Terminologia Sonographica. It proposes the systematic collection of existing descriptors, their translation into operationalizable categories, and consensus-based validation involving ultrasound societies. The aim is to establish an ontologically precise, interoperable semantic infrastructure for quality assurance, digital documentation, and future AI-supported decision-making.