An attention-supervised full-resolution residual network for the segmentation of breast ultrasound images.

PURPOSE Breast cancer is the most common cancer among women worldwide. Medical ultrasound imaging is one of the widely applied breast imaging methods for breast tumors. Automatic breast ultrasound (BUS) image segmentation can measure the size of tumors objectively. However, various ultrasound artifacts hinder segmentation. We proposed an attention-supervised full-resolution residual network (ASFRRN) to segment tumors from BUS images. METHODS In the proposed method, Global Attention Upsample (GAU) and deep supervision were introduced into a full-resolution residual network (FRRN), where GAU learns to merge features at different levels with attention for deep supervision. Two datasets were employed for evaluation. One (Dataset A) consisted of 163 BUS images with tumors (53 malignant and 110 benign) from UDIAT Centre Diagnostic, and the other (Dataset B) included 980 BUS images with tumors (595 malignant and 385 benign) from the Sun Yat-sen University Cancer Center. The tumors from both datasets were manually segmented by medical doctors. For evaluation, the Dice coefficient (Dice), Jaccard similarity coefficient (JSC), and F1 score were calculated. RESULTS For Dataset A, the proposed method achieved higher Dice (84.3 ± 10.0%), JSC (75.2 ± 10.7%), and F1 score (84.3 ± 10.0%) than the previous best method: FRRN. For Dataset B, the proposed method also achieved higher Dice (90.7 ± 13.0%), JSC (83.7 ± 14.8%), and F1 score (90.7 ± 13.0%) than the previous best methods: DeepLabv3 and dual attention network (DANet). For Dataset A+B, the proposed method achieved higher Dice (90.5 ± 13.1%), JSC (83.3 ± 14.8%), and F1 score (90.5 ± 13.1%) than the previous best method: DeepLabv3. Additionally, the parameter number of ASFRRN was only 10.6 M, which is less than those of DANet (71.4 M) and DeepLabv3 (41.3 M). CONCLUSIONS We proposed ASFRRN, which combined with FRRN, attention mechanism, and deep supervision to segment tumors from BUS images. It achieved high segmentation accuracy with a reduced parameter number.

[1]  Jinhua Yu,et al.  Automatic tumor segmentation in breast ultrasound images using a dilated fully convolutional network combined with an active contour model , 2018, Medical physics.

[2]  Bram van Ginneken,et al.  A survey on deep learning in medical image analysis , 2017, Medical Image Anal..

[3]  A. Jemal,et al.  Cancer statistics, 2018 , 2018, CA: a cancer journal for clinicians.

[4]  Reyer Zwiggelaar,et al.  Automated Breast Ultrasound Lesions Detection Using Convolutional Neural Networks , 2018, IEEE Journal of Biomedical and Health Informatics.

[5]  Qinghua Huang,et al.  Breast ultrasound image segmentation: a survey , 2017, International Journal of Computer Assisted Radiology and Surgery.

[6]  Rebecca L. Siegel Mph,et al.  Cancer statistics, 2018 , 2018 .

[7]  Xuelong Li,et al.  Segmentation of breast ultrasound image with semantic classification of superpixels , 2020, Medical Image Anal..

[8]  Seyed-Ahmad Ahmadi,et al.  V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation , 2016, 2016 Fourth International Conference on 3D Vision (3DV).

[9]  M. Giger,et al.  Computerized lesion detection on breast ultrasound. , 2002, Medical physics.

[10]  R. Chang,et al.  Tumor detection in automated breast ultrasound images using quantitative tissue clustering. , 2014, Medical physics.

[11]  Sheng-Fang Huang,et al.  Neural network analysis applied to tumor segmentation on 3D breast ultrasound images , 2008, 2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro.

[12]  Lei Zhu,et al.  Multiscale geodesic active contours for ultrasound image segmentation using speckle reducing anisotropic diffusion , 2014 .

[13]  Seunghoon Hong,et al.  Learning Deconvolution Network for Semantic Segmentation , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[14]  Noriaki Ohuchi,et al.  Adjunctive ultrasonography for breast cancer screening – Authors' reply , 2016, The Lancet.

[15]  Dacheng Tao,et al.  On Combining Biclustering Mining and AdaBoost for Breast Tumor Classification , 2020, IEEE Transactions on Knowledge and Data Engineering.

[16]  Jun Fu,et al.  Dual Attention Network for Scene Segmentation , 2018, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[17]  Bastian Leibe,et al.  Full-Resolution Residual Networks for Semantic Segmentation in Street Scenes , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[18]  Alfonso Reginelli,et al.  Reducing Costs of Breast Examination: Ultrasound Performance and Inter-Observer Variability of Expert Radiologists Versus Residents , 2016, Cancer investigation.

[19]  Dar-Ren Chen,et al.  Watershed segmentation for breast tumor in 2-D sonography. , 2004, Ultrasound in medicine & biology.

[20]  Enhua Wu,et al.  Squeeze-and-Excitation Networks , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[21]  Hongzhi Wang,et al.  Fast anatomy segmentation by combining coarse scale multi-atlas label fusion with fine scale corrective learning , 2018, Comput. Medical Imaging Graph..

[22]  Fei Xu,et al.  Automatic Breast Ultrasound Image Segmentation: A Survey , 2017, Pattern Recognit..

[23]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[24]  Vijayan K. Asari,et al.  Nuclei Segmentation with Recurrent Residual Convolutional Neural Networks based U-Net (R2U-Net) , 2018, NAECON 2018 - IEEE National Aerospace and Electronics Conference.

[25]  Xuelong Li,et al.  Graph-based learning for segmentation of 3D ultrasound images , 2015, Neurocomputing.

[26]  Joseph O. Deasy,et al.  Multiple Resolution Residually Connected Feature Streams for Automatic Lung Tumor Segmentation From CT Images , 2019, IEEE Transactions on Medical Imaging.

[27]  Xuelong Li,et al.  Optimized graph-based segmentation for ultrasound images , 2014, Neurocomputing.

[28]  Trevor Darrell,et al.  Fully Convolutional Networks for Semantic Segmentation , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[29]  Lian-Wen Jin,et al.  A robust graph-based segmentation method for breast tumors in ultrasound images. , 2012, Ultrasonics.

[30]  Dimitris N. Metaxas,et al.  Combining low-, high-level and empirical domain knowledge for automated segmentation of ultrasonic breast lesions , 2003, IEEE Transactions on Medical Imaging.

[31]  T. Endo,et al.  Sensitivity and specificity of mammography and adjunctive ultrasonography to screen for breast cancer in the Japan Strategic Anti-cancer Randomized Trial (J-START): a randomised controlled trial , 2016, The Lancet.

[32]  Roberta A. Jong,et al.  Ultrasound as the Primary Screening Test for Breast Cancer: Analysis From ACRIN 6666. , 2016, Journal of the National Cancer Institute.

[33]  Kaiming He,et al.  Feature Pyramid Networks for Object Detection , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[34]  Navin Kumar Manaswi,et al.  Understanding and Working with Keras , 2018 .