Fifty-seventh annual meeting of the American association of physicists in medicine
SU-E-I-76: A New Approach for Evaluating Visibility of Static Image Displayed On Liquid-Crystal Display by Using Detection Time
A receiver operating characteristics analysis is one of the most useful methods for evaluating diagnostic accuracy. However, methods to evaluate the visibility of images displayed on a liquid-crystal display (LCD) are not well established. The detection time of signals can be used to evaluate visibility. However, detection time depends on the characteristics each observer's eye-brain system. This study aimed to evaluate the visibility of static images with a normalized-rank approach of detection times.
Sixty images with 2–6 bright or dark signals against background luminance on an LCD were observed by 10 observers. We measured the time required for the observer to determine whether signals existed just after the image was displayed on the LCD. This study assumed that images with short detectability times would be superior to those with longer times. The detection times of the 60 images were sorted for each observer for additional analyses of visibility. A normalized-rank approach was applied to the ranked data in order to evaluate the visibility of bright and dark signals. In addition, we calculated the positive predictive value (PPV) and the negative predictive value (NPV) for all observers.
In the normalized-rank approach, the visibility of dark signals against background luminance was statistically superior to that of bright signals (p = 0.0393). The PPV/NPV values for the bright and dark signals were 98%/96% and 97%/97%, respectively.
This novel normalized-rank approach for analyzing detection time can be used to evaluate the visibility of simple signals on an LCD and compare them with PPV and NPV.