Detection of architectural distortion from the ridges in a digitized mammogram

Reconstruction of three‐dimensional linear structures in the breast from craniocaudal and mediolateral oblique mammographic views

“…The 2D projections of the linear structures of our synthetic pattern on two orthogonal planes were taken as CC and MLO projections of the 3D linear structures. The statistical ridge detector [14] was applied on these synthetic CC and MLO images. After applying connected component labelling on the output of the statistical ridge detector on these CC and MLO images, Algorithms 3 and 4 were applied on the connected components for 3D reconstruction.…”

“…9 c and d can be easily identified (as highlighted by the arrows). Based on the length and variation in ridge intensity (the output of the statistical ridge detector [14] on Figs. 9 a and b ) of a connected component 11 structures in the CC view were selected manually as the corresponding structures of W .…”

“… Illustration of the extraction of linear structures from a mammogram (a) Section of an MLO view of a mammogram from the DDSM database (File: 0006_LEFT_MLO.pgm) [13], (b) Ridges extracted from (a) by a statistical ridge detector [14], (c) Grouping of pixels in (b) into linear structures based on connected component labelling. Pixels lying on the same linear structure have been assigned the same colour…”

A New Method for Detecting Architectural Distortion in Mammograms by NonSubsampled Contourlet Transform and Improved PCNN

“…The results are shown in Table 8. In Table 8, it shows that the sensitivity of the proposed method is lower than that of the four methods, but the specificity and accuracy are higher than those proposed in [18], and the AUC is higher than those proposed in [17,19,20]. The sensitivity is lower, because as the image density rating increases, between the contrast of the AD area and the contrast of the background area becomes smaller and smaller, resulting in a decrease in the number of TP, which in turn causes a decrease in sensitivity.…”

“…Narváez et al extracted the linear structure information in the ROI and the edge, and formed a new feature vector according to different weights, AD was detected by SVM, and with an accuracy of 89% [18]. Akhtar et al proposed using a radial ridge to detect AD based on the linear characteristics, with a sensitivity of 85% and with a specificity of 80% [19]. Costa et al proposed using deep learning to detect architectural distortion in mammograms, however, due to the limitation of the dataset size, the accuracy is only 86.1% [20].…”