Dana S Al Mousa1, Patrick C Brennan2, Elaine A Ryan2, Warwick B Lee3, Jennifer Tan4, Claudia Mello-Thoms2. 1. Medical Image Optimisation and Perception Group (MIOPeG), Discipline of Medical Imaging and Radiation Sciences, Faculty of Health Sciences, University of Sydney, Cumberland Campus, East St, P.O. Box 170, Lidcombe, New South Wales 2141, Australia. Electronic address: dalm3874@uni.sydney.edu.au. 2. Medical Image Optimisation and Perception Group (MIOPeG), Discipline of Medical Imaging and Radiation Sciences, Faculty of Health Sciences, University of Sydney, Cumberland Campus, East St, P.O. Box 170, Lidcombe, New South Wales 2141, Australia. 3. BreastScreen, Cancer Institute of NSW, Sydney, New South Wales, Australia. 4. Sydney South West Area Health Services, Sydney, New South Wales, Australia.
Abstract
RATIONALE AND OBJECTIVES: To determine the impact of mammographic breast density on the visual search process of radiologists when reading digital mammograms. MATERIALS AND METHODS: Institutional review board approval was obtained. A set of 149 craniocaudal digital mammograms were read by seven radiologists, and observer search patterns were recorded. Total time examining each case, time to first hit the lesion, dwell time, and number of hits per area were calculated. The nonparametric Mann-Whitney U test was used for statistical evaluation. RESULTS: In both low- and high-mammographic density cases, significant increases were observed in the time to first hit lesions when they were located outside, compared to overlying fibroglandular dense tissue (P = .001). Significantly longer dwell time (P = .003) and greater number of fixations (P = .0003) were observed when the lesions were situated within--rather than outside--the dense fibroglandular tissue. CONCLUSIONS: Increased mammographic breast density changes radiologists' visual search patterns. Dense areas of the parenchyma attracted greater visual attention in both high- and low-mammographic density cases, resulting in faster detection of lesions overlying the fibroglandular dense tissue, along with longer dwell times and greater number of fixations, as compared to lesions located outside the dense fibroglandular regions.
RATIONALE AND OBJECTIVES: To determine the impact of mammographic breast density on the visual search process of radiologists when reading digital mammograms. MATERIALS AND METHODS: Institutional review board approval was obtained. A set of 149 craniocaudal digital mammograms were read by seven radiologists, and observer search patterns were recorded. Total time examining each case, time to first hit the lesion, dwell time, and number of hits per area were calculated. The nonparametric Mann-Whitney U test was used for statistical evaluation. RESULTS: In both low- and high-mammographic density cases, significant increases were observed in the time to first hit lesions when they were located outside, compared to overlying fibroglandular dense tissue (P = .001). Significantly longer dwell time (P = .003) and greater number of fixations (P = .0003) were observed when the lesions were situated within--rather than outside--the dense fibroglandular tissue. CONCLUSIONS: Increased mammographic breast density changes radiologists' visual search patterns. Dense areas of the parenchyma attracted greater visual attention in both high- and low-mammographic density cases, resulting in faster detection of lesions overlying the fibroglandular dense tissue, along with longer dwell times and greater number of fixations, as compared to lesions located outside the dense fibroglandular regions.