PeMNet for Pectoral Muscle Segmentation

Abstract

As an important imaging modality, mammography is considered to be the global gold standard for early detection of breast cancer. Computer-Aided (CAD) systems have played a crucial role in facilitating quicker diagnostic procedures, which otherwise could take weeks if only radiologists were involved. In some of these CAD systems, breast pectoral segmentation is required for breast region partition from breast pectoral muscle for specific analysis tasks. Therefore, accurate and efficient breast pectoral muscle segmentation frameworks are in high demand. Here, we proposed a novel deep learning framework, which we code-named PeMNet, for breast pectoral muscle segmentation in mammography images. In the proposed PeMNet, we integrated a novel attention module called the Global Channel Attention Module (GCAM), which can effectively improve the segmentation performance of Deeplabv3+ using minimal parameter overheads. In GCAM, channel attention maps (CAMs) are first extracted by concatenating feature maps after paralleled global average pooling and global maximum pooling operation. CAMs are then refined and scaled up by multi-layer perceptron (MLP) for elementwise multiplication with CAMs in next feature level. By iteratively repeating this procedure, the global CAMs (GCAMs) are then formed and multiplied elementwise with final feature maps to lead to final segmentation. By doing so, CAMs in early stages of a deep convolution network can be effectively passed on to later stages of the network and therefore leads to better information usage. The experiments on a merged dataset derived from two datasets, INbreast and OPTIMAM, showed that PeMNet greatly outperformed state-of-the-art methods by achieving an IoU of 97.46%, global pixel accuracy of 99.48%, Dice similarity coefficient of 96.30%, and Jaccard of 93.33%, respectively.

Comparison between two packages for pectoral muscle removal on mammographic images

“…As a matter of fact, important issues with deep models for pectoral removal are the robustness of the methods and the training phase. Before the advent of deep learning, feature-based methods dominated the field [34][35][36][37]. The robustness of these kinds of systems remains to be improved as variations in the images could lead to wrong removal.…”

A New Fat-Removal-Based Preprocessing Pipeline for MLO View in Digital Mammograms

“…Finally, methods based on learning features are based on artificial neural networks [14] and deep neural networks [21], [29], [30]. Some of the most relevant are based on semantic segmentation [31], [32], [33], [34], [35]. One of the main disadvantages of these techniques is the imbalance of classes to be segmented.…”

Wavelet-Based Automatic Pectoral Muscle Segmentation for Mammograms

“…Machine learning methodologies are also u4lized for the segmenta4on of the pectoral muscle from mammograms [26,[32][33][34]. An example of this is seen through a connected component labeling method to remove the pectoral muscle region of the breast from the mammogram.…”