Geometry-Based Pectoral Muscle Segmentation From MLO Mammogram Views

Abstract

Computer-aided diagnosis systems (CADx) play a major role in the early diagnosis of breast cancer. Extracting the breast region precisely from a mammogram is an essential component of CADx for mammography. The appearance of the pectoral muscle on medio-lateral oblique (MLO) views increases the false positive rate in CADx. Therefore, the pectoral muscle should be identified and removed from the breast region in an MLO image before further analysis. None of the previous pectoral muscle segmentation methods address all breast types based on the breast imaging-reporting and data system tissue density classes. In this paper, we deal with this deficiency by introducing a new simple yet effective method that combines geometric rules with a region growing algorithm to support the segmentation of all types of pectoral muscles (normal, convex, concave, and combinatorial). Experimental segmentation accuracy results were reported for four tissue density classes on 872 MLO images from three publicly available datasets. An average Jaccard index and Dice similarity coefficient of 0.972 ± 0.003 and 0.985 ± 0.001 were obtained, respectively. The mean Hausdorff distance between the contours detected by our method and the ground truth is below 5 mm for all datasets. An average acceptable segmentation rate of ∼95% was achieved outperforming several state-of-the-art competing methods. Excellent results were obtained even for the most challenging class of extremely dense breasts.Computer-aided diagnosis systems (CADx) play a major role in the early diagnosis of breast cancer. Extracting the breast region precisely from a mammogram is an essential component of CADx for mammography. The appearance of the pectoral muscle on medio-lateral oblique (MLO) views increases the false positive rate in CADx. Therefore, the pectoral muscle should be identified and removed from the breast region in an MLO image before further analysis. None of the previous pectoral muscle segmentation methods address all breast types based on the breast imaging-reporting and data system tissue density classes. In this paper, we deal with this deficiency by introducing a new simple yet effective method that combines geometric rules with a region growing algorithm to support the segmentation of all types of pectoral muscles (normal, convex, concave, and combinatorial). Experimental segmentation accuracy results were reported for four tissue density classes on 872 MLO images from three publicly available datasets. An average Jaccard index and Dice similarity coefficient of 0.972 ± 0.003 and 0.985 ± 0.001 were obtained, respectively. The mean Hausdorff distance between the contours detected by our method and the ground truth is below 5 mm for all datasets. An average acceptable segmentation rate of ∼95% was achieved outperforming several state-of-the-art competing methods. Excellent results were obtained even for the most challenging class of extremely dense breasts.

Computer-Aided Detection and Diagnosis of Breast Cancer: a Review

“…In the literature, a series of studies used this dataset to evaluate the performance of ROI segmentation and feature extraction techniques (Taghanaki et al, 2017;Baccouche et al, 2021;Li et al, 2021;Zhang et al, 2020;Alkhaleefah et al, 2022).…”

“…The performance of the proposal was compared based on the Jaccard index and Hausdorff distance with three advanced segmentation techniques for mini MIAS and DDSM datasets and was found to perform better for 79.50% and 74.22% of images. Taghanaki et al (2017) proposed a BI-RADS tissue density-based technique to segment the pectoral muscle from mammograms. The region-growing algorithm merged with geometric rules to identify the pectoral muscle (concave, convex, normal and combinatorial).…”

PeMNet for Pectoral Muscle Segmentation

“…The segmentation finishes when no more pixels can be included [ 13 ]. In another region growing-based method [ 14 ], image intensity is rescaled from 0 to 1 while a classical image contrast enhancement method called CLAHE was used to improve the image contrast. The images were then binarized into binary images using a threshold value of 0.03.…”