A deep learning system to obtain the optimal parameters for a threshold-based breast and dense tissue segmentation

BACKGROUND AND OBJECTIVE Breast cancer is the most frequent cancer in women. The Spanish healthcare network established population-based screening programs in all Autonomous Communities, where mammograms of asymptomatic women are taken with early diagnosis purposes. Breast density assessed from digital mammograms is a biomarker known to be related to a higher risk to develop breast cancer.It is thus crucial to provide a reliable method to measure breast density from mammograms. Furthermore the complete automation of this segmentation process is becoming fundamental as the amount of mammograms increases every day. Important challenges are related with the differences in images from different devices and the lack of an objective gold standard.This paper presents a fully automated framework based on deep learning to estimate the breast density. The framework covers breast detection, pectoral muscle exclusion, and fibroglandular tissue segmentation. METHODS A multi-center study, composed of 1785 women whose "for presentation" mammograms were segmented by two experienced radiologists. A total of 4992 of the 6680 mammograms were used as training corpus and the remaining (1688) formed the test corpus. This paper presents a histogram normalization step that smoothed the difference between acquisition, a regression architecture that learned segmentation parameters as intrinsic image features and a loss function based on the DICE score. RESULTS The results obtained indicate that the level of concordance (DICE score) reached by the two radiologists (0.77) was also achieved by the automated framework when it was compared to the closest breast segmentation from the radiologists. For the acquired with the highest quality device, the DICE score per acquisition device reached 0.84, while the concordance between radiologists was 0.76. CONCLUSIONS An automatic breast density estimator based on deep learning exhibits similar performance when compared with two experienced radiologists. It suggests that this system could be used to support radiologists to ease its work.

[1]  Jack Cuzick,et al.  Clinical and epidemiological issues in mammographic density , 2012, Nature Reviews Clinical Oncology.

[2]  Anil K. Jain.  Data clustering: 50 years beyond K-means , 2010, Pattern Recognit. Lett..

[3]  Pascal Vincent,et al.  Representation Learning: A Review and New Perspectives , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4]  Robert Marti,et al.  A Novel Breast Tissue Density Classification Methodology , 2008, IEEE Transactions on Information Technology in Biomedicine.

[5]  J. James,et al.  The current status of digital mammography. , 2004, Clinical radiology.

[6]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[7]  Brent Mombourquette,et al.  A multi-site study of a breast density deep learning model for full-field digital mammography and digital breast tomosynthesis exams , 2020, ArXiv.

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

[9]  Amr Badr,et al.  Lung Cancer Detection and Classification with 3D Convolutional Neural Network (3D-CNN) , 2017 .

[10]  Nico Karssemeijer,et al.  Unsupervised Deep Learning Applied to Breast Density Segmentation and Mammographic Risk Scoring , 2016, IEEE Transactions on Medical Imaging.

[11]  Carlos Sáez,et al.  Comparative study of probability distribution distances to define a metric for the stability of multi-source biomedical research data , 2013, 2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[12]  Chuan Zhou,et al.  Deep Learning in Medical Image Analysis. , 2017, Advances in experimental medicine and biology.

[13]  Brendan J. Frey,et al.  Deep learning of the tissue-regulated splicing code , 2014, Bioinform..

[14]  Rafael Llobet,et al.  Global parenchymal texture features based on histograms of oriented gradients improve cancer development risk estimation from healthy breasts , 2019, Comput. Methods Programs Biomed..

[15]  Xiaohui Peng,et al.  Deep Learning for Sensor-based Activity Recognition: A Survey , 2017, Pattern Recognit. Lett..

[16]  L. R. Dice.  Measures of the Amount of Ecologic Association Between Species , 1945 .

[17]  Srinivas C. Turaga,et al.  Connectomic reconstruction of the inner plexiform layer in the mouse retina , 2013, Nature.

[18]  Christopher Y. Park,et al.  Whole-genome deep-learning analysis identifies contribution of noncoding mutations to autism risk , 2019, Nature Genetics.

[19]  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).

[20]  Ming Yang,et al.  DeepFace: Closing the Gap to Human-Level Performance in Face Verification , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[21]  Rafael Wiemker,et al.  Inter- and intra-observer variations in the delineation of lesions in mammograms , 2015, Medical Imaging.

[22]  Sung Hun Kim,et al.  Interpretive Performance and Inter-Observer Agreement on Digital Mammography Test Sets , 2018, Korean journal of radiology.

[23]  Reyer Zwiggelaar,et al.  Mammographic Segmentation and Density Classification: A Fractal Inspired Approach , 2016, Digital Mammography / IWDM.

[24]  Jonathan Tompson,et al.  Efficient object localization using Convolutional Networks , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[25]  A. Paterson,et al.  Mammographic breast density as an intermediate phenotype for breast cancer. , 2005, The Lancet. Oncology.

[26]  Mireille J. M. Broeders,et al.  Comparing Visually Assessed BI-RADS Breast Density and Automated Volumetric Breast Density Software: A Cross-Sectional Study in a Breast Cancer Screening Setting , 2015, PloS one.

[27]  Tara N. Sainath,et al.  Deep Neural Networks for Acoustic Modeling in Speech Recognition , 2012 .

[28]  Christiane K. Kuhl,et al.  The Changing World of Breast Cancer: A Radiologist's Perspective , 2016, Plastic surgical nursing : official journal of the American Society of Plastic and Reconstructive Surgical Nurses.

[29]  Xiang Zhang,et al.  OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks , 2013, ICLR.

[30]  H. Sebastian Seung,et al.  Convolutional nets for reconstructing neural circuits from brain images acquired by serial section electron microscopy , 2019, Current Opinion in Neurobiology.

[31]  Nan Wu,et al.  Breast Density Classification with Deep Convolutional Neural Networks , 2017, 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[32]  Rafael Llobet,et al.  Semi-automated and fully automated mammographic density measurement and breast cancer risk prediction , 2014, Comput. Methods Programs Biomed..

[33]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[34]  Rafael Llobet,et al.  Validation of DM-Scan, a computer-assisted tool to assess mammographic density in full-field digital mammograms , 2013, SpringerPlus.

[35]  S. Ciatto,et al.  Categorizing breast mammographic density: intra- and interobserver reproducibility of BI-RADS density categories. , 2005, Breast.

[36]  Xiao Zhao,et al.  The connected-component labeling problem: A review of state-of-the-art algorithms , 2017, Pattern Recognit..

[37]  Carlos Sáez,et al.  Stability metrics for multi-source biomedical data based on simplicial projections from probability distribution distances , 2017, Statistical methods in medical research.

[38]  Kesheng Wu,et al.  Optimizing two-pass connected-component labeling algorithms , 2009, Pattern Analysis and Applications.

[39]  Fei Wang,et al.  Deep learning for healthcare: review, opportunities and challenges , 2018, Briefings Bioinform..

[40]  Shiju Yan,et al.  A robust method for segmenting pectoral muscle in mediolateral oblique (MLO) mammograms , 2018, International Journal of Computer Assisted Radiology and Surgery.

[41]  Ke Zhou,et al.  Automatic Pectoral Muscle Region Segmentation in Mammograms Using Genetic Algorithm and Morphological Selection , 2018, Journal of Digital Imaging.

[42]  Krzysztof J. Geras,et al.  Artificial Intelligence for Mammography and Digital Breast Tomosynthesis: Current Concepts and Future Perspectives. , 2019, Radiology.

[43]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[44]  R. Nishikawa,et al.  Automated mammographic breast density estimation using a fully convolutional network , 2018, Medical physics.

[45]  L. Liberman,et al.  Breast imaging reporting and data system (BI-RADS). , 2002, Radiologic clinics of North America.

[46]  Dinggang Shen,et al.  Scalable High-Performance Image Registration Framework by Unsupervised Deep Feature Representations Learning , 2016, IEEE Transactions on Biomedical Engineering.

[47]  Rekha Lakshmanan,et al.  Pectoral Muscle Boundary Detection in Mammograms Using Homogeneous Contours , 2015, 2015 Fifth International Conference on Advances in Computing and Communications (ICACC).

[48]  B. Thirumala Rao,et al.  Novel Approach to Segment the Pectoral Muscle in the Mammograms , 2019 .

[49]  P. Skaane.  Studies comparing screen-film mammography and full-field digital mammography in breast cancer screening: Updated review , 2009, Acta radiologica.

[50]  R. Barzilay,et al.  Mammographic Breast Density Assessment Using Deep Learning: Clinical Implementation. , 2019, Radiology.