Mass Detection and Segmentation in Digital Breast Tomosynthesis Using 3D-Mask Region-Based Convolutional Neural Network: A Comparative Analysis.

Ming Fan,Lihua Li,Xin Gao,Yajia Gu,Weijun Peng,Huizhong Zheng,Chao You,Shuo Zheng

doi:10.3389/fmolb.2020.599333

Abstract

Digital breast tomosynthesis (DBT) is an emerging breast cancer screening and diagnostic modality that uses quasi-three-dimensional breast images to provide detailed assessments of the dense tissue within the breast. In this study, a framework of a 3D-Mask region-based convolutional neural network (3D-Mask RCNN) computer-aided diagnosis (CAD) system was developed for mass detection and segmentation with a comparative analysis of performance on patient subgroups with different clinicopathological characteristics. To this end, 364 samples of DBT data were used and separated into a training dataset (n = 201) and a testing dataset (n = 163). The detection and segmentation results were evaluated on the testing set and on subgroups of patients with different characteristics, including different age ranges, lesion sizes, histological types, lesion shapes and breast densities. The results of our 3D-Mask RCNN framework were compared with those of the 2D-Mask RCNN and Faster RCNN methods. For lesion-based mass detection, the sensitivity of 3D-Mask RCNN-based CAD was 90% with 0.8 false positives (FPs) per lesion, whereas the sensitivity of the 2D-Mask RCNN- and Faster RCNN-based CAD was 90% at 1.3 and 2.37 FPs/lesion, respectively. For breast-based mass detection, the 3D-Mask RCNN generated a sensitivity of 90% at 0.83 FPs/breast, and this framework is better than the 2D-Mask RCNN and Faster RCNN, which generated a sensitivity of 90% with 1.24 and 2.38 FPs/breast, respectively. Additionally, the 3D-Mask RCNN achieved significantly (p < 0.05) better performance than the 2D methods on subgroups of samples with characteristics of ages ranged from 40 to 49 years, malignant tumors, spiculate and irregular masses and dense breast, respectively. Lesion segmentation using the 3D-Mask RCNN achieved an average precision (AP) of 0.934 and a false negative rate (FNR) of 0.053, which are better than those achieved by the 2D methods. The results suggest that the 3D-Mask RCNN CAD framework has advantages over 2D-based mass detection on both the whole data and subgroups with different characteristics.

Highlights

Breast cancer is the most common malignancy in women
The combination of Digital breast tomosynthesis (DBT) and mammography resulted in significant gains in the sensitivity and specificity of cancer detection compared with digital mammography (DM) alone (Fontaine et al, 2019; Li et al, 2019; Skaane et al, 2019)
The results of the 3D-Mask RCNN computeraided diagnosis (CAD) system achieved a sensitivity of 90% with 0.80 false positives (FPs)/lesion and 0.83 FPs/breast, respectively

Summary

Introduction

Breast cancer is the most common malignancy in women. Full-field digital mammography (FFDM) is commonly used to screen for breast cancer (Nystrom et al, 2002). Mammography has an inherent limitation when tissue overlaps, especially in dense breasts, which causes mammography to miss some suspicious cancerous lesions (Carney et al, 2003). Digital breast tomosynthesis (DBT) is an emerging breast cancer screening and diagnostic modality that takes quasi-threedimensional imaging that can be used to provide a detailed assessment of the dense tissue within the breast. DBT has decreased the effect of overlapping tissue on screening, thereby improving lesion detection, characterization and diagnosis and making this approach superior to digital mammography (DM) (Michell et al, 2012; Haas et al, 2013). The integration of DBT into the diagnostic setting is associated with improved diagnostic performance of breast cancer due to the increased specificity (Bahl et al, 2019; Conant et al, 2019). Due to its improvements in patient diagnosis efficiency, DBT is becoming the standard of care in both screening and diagnostic breast images (Chong et al, 2019)

Methods

Results

Conclusion