DEVELOPMENT AND CLINICAL VALIDATION OF A SOFTWARE FOR MENSARY DENSITY MEASUREMENT
MAMMARY DENSITY; MAMOGRAPHY; BREAST CANCER; MAMMARY DENSITY MEASUREMENT; SOFTWARE
Breast density has assumed increasing importance due to its relation with breast cancer, but its measurement has limitations linked to the subjectivity of visual validation or difficult access to automatic methods of volumetric measurement.
Objective: Development and clinical validation of the computer program to determine mammographic density in an objective and reproducible way from flat mammography images.
Methods: After segmentation of the image, separating the background of the breast image, the operator positions the cursor that limits the region of interest on the pectoral muscle in the MLO incidence. In CC incidence, the image is standardized based on the optical density of the pectoral muscle and the proportion of dense tissue is calculated by the program. For standardization, 282 mammograms were used and for clinical validation the mammograms obtained in 2 different times in 200 women. The program was developed in MatLab.
Results: The intraclass correlation coefficient (ICC) between the measurements of breast density by the program and by a radiologist was 0.88, with a bias of only 1.37 percentage points and a 95% concordance limit of 13.4 points percentages; in the evaluation of the interobserver reliability, with two radiologists with different experience, the CCI was 0.87; in the assessment of intraobserver reliability, the ICC was 0.92. The distribution among the density classes is close to the values obtained with the volumetric software.
Conclusion: The determination of the breast density in the Aguida program in flat mammography imaging showed high agreement with the visual determination by radiologist, high inter-and intra-observer reliability.