Efron Nathan, Chaudry Aysha
Institute of Health and Biomedical Innovation and School of Optometry, Queensland University of Technology, Queensland, Australia.
Clin Exp Optom. 2007 Sep;90(5):361-6. doi: 10.1111/j.1444-0938.2007.00171.x.
This study was conducted to investigate grading performance when estimating the severity of static versus dynamic images of contact lens-related ocular pathology.
Thirty-eight subjects used the Efron Grading Scales for Contact Lens Complications to grade the severity of ocular pathological changes depicted in static and dynamic (movie clip) computer-displayed images of each of the following contact lens complications: bulbar conjunctival redness, limbal redness, papillary conjunctivitis, corneal staining, corneal infiltrates and meibomian gland dysfunction. The viewing of static and dynamic images was separated by seven weeks.
Grades assigned to dynamic images were 0.6 and 0.7 grading scale units higher than those assigned to static images for limbal redness and papillary conjunctivitis, respectively (p < 0.0001 for both). No difference was observed for the other four complications. There was an apparent trend for grading variability to be reduced (that is, observers grading in closer agreement) when grading dynamic versus static images.
Absolute grades based on an assessment of signs of pathology represented in static images may, in some instances, underestimate the true severity of the condition.
本研究旨在调查评估与隐形眼镜相关的眼部病变静态图像和动态图像严重程度时的分级表现。
38名受试者使用隐形眼镜并发症的埃弗龙分级量表，对以下每种隐形眼镜并发症的静态和动态（电影片段）计算机显示图像中所描绘的眼部病理变化严重程度进行分级：球结膜充血、角膜缘充血、乳头性结膜炎、角膜染色、角膜浸润和睑板腺功能障碍。静态图像和动态图像的查看间隔7周。
角膜缘充血和乳头性结膜炎的动态图像分级分别比静态图像高0.6和0.7个分级量表单位（两者p均<0.0001）。其他四种并发症未观察到差异。在对动态图像与静态图像进行分级时，分级变异性有明显降低的趋势（即观察者分级的一致性更高）。
基于对静态图像中病理体征评估得出的绝对分级在某些情况下可能低估病情的真实严重程度。
