Keywords:
Multicentre study, Observational, Not applicable, Quality assurance, Comparative studies, Digital radiography, Thorax, Professional issues, Artificial Intelligence, Chest
Authors:
M. Englmaier1, D. Sasse1, D. Pfeiffer1, M. Kotnik2, L. Lin2, H. J. Lamb2, J. Conradsen3, J. Fløtten3, N. Wieberneit4; 1Munich/DE, 2Leiden/NL, 3Herning/DK, 4Hamburg/DE
DOI:
10.26044/ecr2020/C-05601
Purpose
In radiology, good image quality is the prerequisite for accurate diagnosis and appropriate treatment decisions. While chest radiography is one of the most frequently performed imaging procedures [1], achieving consistently optimal image quality continues to be a challenge [2,3]. The underlying reasons for this are manifold, but include declining education, changing communication behavior due to digitalization and time-constraints [4]. At the same time, perceived image quality differs quite significantly among individual observers [5,6]. Among the observed image quality shortcomings, positioning errors are the most frequently encountered [7].
In order to address inadequate image quality, an AI-based positioning quality check was developed [8] to enable feedback to the technologist.
The purpose of this work is to investigate the inter-rater agreement in the assessment of positioning quality of chest x-ray images in the context of establishing thresholds between good and inadequate image quality for the AI-based positioning quality check.