MODELING HUMAN EYE BEHAVIOR DURING MAMMOGRAPHIC SCANNING - PRELIMINARY-RESULTS

Understanding how people acquire information from pictures-radiographs , maps, charts, photographs, drawings, and other static images-can be an important component in understanding, aiding, and eventually automa ting a wide range of diagnostic tasks, In the experiment reported here , we investigate the use of an inexpensive and unobtrusive eye trackin g system to explore relationships between visual scanning patterns, pu pillary response, and the clinical diagnoses of mammographic experts, One radiologist and three radiological technicians each examined a ser ies of 14 mammograms for indications of abnormalities associated with breast cancer, The status of each mammogram was verified by biopsy, Th e eye-tracking system was used to measure and record eye position and pupil diameter as a function of time as the subjects scanned the mammo grams. Three treatments were applied to the scan data to model the exp erts' eve behaviors, These included quantification of dwell time and p upil diameter as a function diagnostic accuracy in regions of the mamm ogram where abnormalities existed or were perceived; independent clust ering of lookpoints without respect to abnormalities: and analysis of scan transitions between lookpoint clusters, Results of the analysis w ere consistent with extensive prior studies of eye-scan measures recor ded during the diagnosis of abnormalities on chest radiograms. This pr eliminary investigation pro,ides a proof of concept for use of the eye -tracking technology, experimental protocols, and analysis methodologi es as the basis for expanded mammographic studies, with the promise of eventual adaptation as a source of diagnostic information in clinical practice.