作者: Feng Li , Seiji Kumazawa , Junji Shiraishi , Qiang Li , Roger Engelmann
DOI: 10.1016/J.ACRA.2008.10.016
关键词:
摘要: Rationale and Objectives The aim of this study was to investigate the subjective similarity for pairs images with various abnormal patterns diffuse interstitial lung disease on thin-section computed tomography by experienced radiologists explore a basis selecting similar assist radiologists' interpretation. Materials Methods Four major (ground-glass opacity, nodular reticular honeycombing) tomographic were identified at least two three radiologists. One radiologist manually selected 104 image pairs, in which each pair had same pattern appearance. An additional 208 randomly evenly divided among four patterns. These then rated (on continuous scale ranging from 0 = not all 1.0 almost identical) 12 Results For radiologist-selected mean 0.72. higher (0.47) than varying (0.27) (P Conclusion Subjective ratings can be measured reliably reproducibly will provide selection