Impact of visual fatigue on observer performance

Our overall hypothesis is that current radiology practice produces oculomotor fatigue reducing diagnostic accuracy. The goal of this study is to determine whether accommodative stability and diagnostic accuracy are reduced following digital radiology interpretation. We are collecting data at two points in time - once in the morning prior to diagnostic reading and once in the afternoon after reading. Subjects are completing surveys about their current physical status and number of hours spent reading that day along and the type of images read. We are measuring accommodation using the WAM- 5500 Auto Refkeratometer. Subjects view bone images with subtle fractures and dislocations to determine if a fracture is present, locate it, and provide rating of their decision confidence to be used in a ROC analysis of the data. Preliminary results confirm our previous findings that we can measure visual fatigue. Radiologists are less able to focus on a distinct point, especially at near distances, after a day of reading images on digital displays as opposed to before any reading takes place. The SOFI and SSQ measures also indicate that radiologists are more fatigued at the end of a day's reading as compared to before. The confidence ratings are being evaluated using ROC techniques. The results so far suggest a reduction in diagnostic accuracy with tired eyes. Preliminary data from measuring visual accommodation and observer performance support our hypothesis that radiologists suffer visual fatigue after a day reading diagnostic images from digital displays reducing interpretation accuracy.