SGER: Data Collection for an Eye Behavior Model of Mammographers

点击翻译按钮获取中文摘要

9803735 Mayer The goal of this proposed research is to determine whether the eye behavior of mammographers is correlated to their diagnostic accuracy. A clear understanding of the relationship between diagnostic accuracy and eye behavior has the potential to save many of the lives of the estimated 180,200 woman afflicted with breast cancer each year. The information from studying the visual processes of mammographers can be used to develop effective training methods and computer aided diagnostic algorithms to reduce the occurrence of false positives and false negatives associated with this disease. To test the research hypothesis, an unobtrusive PC based eye-gaze system, called ERICA, will record the visual pattern that mammographers utilize while screening mammograms. The ERICA hardware, positioned beneath a lightbox, can capture the visual process a reader uses in mammographic screening. The ERICA system captures eye position and pupil diameter as a function of time. Mammographic experts from the University of Virginia Health Sciences Center will screen 200 randomly selected mammograms over the course of a year, Cases will be selected from an existing telemammography project at the University of Virginia, which includes 400 mammograms. ***