{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,9,8]],"date-time":"2025-09-08T05:34:21Z","timestamp":1757309661655,"version":"3.41.2"},"reference-count":47,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2022,11,24]],"date-time":"2022-11-24T00:00:00Z","timestamp":1669248000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100002855","name":"Ministry of Science and Technology of the People's Republic of China","doi-asserted-by":"publisher","award":["2017YFC0112800"],"award-info":[{"award-number":["2017YFC0112800"]}],"id":[{"id":"10.13039\/501100002855","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100008838","name":"Shanghai Municipal Commission of Economy and Informatization","doi-asserted-by":"publisher","award":["XX-XXFZ-02-17-4126"],"award-info":[{"award-number":["XX-XXFZ-02-17-4126"]}],"id":[{"id":"10.13039\/501100008838","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100003399","name":"Science and Technology Commission of Shanghai Munixcipality","doi-asserted-by":"publisher","award":["17411952300"],"award-info":[{"award-number":["17411952300"]}],"id":[{"id":"10.13039\/501100003399","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Comput. Sci."],"abstract":"<jats:sec><jats:title>Introduction<\/jats:title><jats:p>Early breast carcinomas can be effectively diagnosed and controlled. However, it demands extra work and radiologist in China often suffer from overtime working due to too many patients, even experienced ones could make mistakes after overloaded work. To improve the efficiency and reduce the rate of misdiagnosis, automatic breast diagnosis on Magnetic Resonance Imaging (MRI) images is vital yet challenging for breast disease screening and successful treatment planning. There are some obstacles that hinder the development of automatic approaches, such as class-imbalance of samples, hard mimics of lesions, etc. In this paper, we propose a coarse-to-fine algorithm to address those problems of automatic breast diagnosis on multi-series MRI images. The algorithm utilizes deep learning techniques to provide breast segmentation, tumor segmentation and tumor classification functions, thus supporting doctors' decisions in clinical practice.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>In proposed algorithm, a DenseUNet is firstly employed to extract breast-related regions by removing irrelevant parts in the thoracic cavity. Then, by taking advantage of the attention mechanism and the focal loss, a novel network named Attention Dense UNet (ADUNet) is designed for the tumor segmentation. Particularly, the focal loss in ADUNet addresses class-imbalance and model overwhelmed problems. Finally, a customized network is developed for the tumor classification. Besides, while most approaches only consider one or two series, the proposed algorithm takes in account multiple series of MRI images.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>Extensive experiments are carried out to evaluate its performance on 435 multi-series MRI volumes from 87 patients collected from Tongji Hospital. In the dataset, all cases are with benign, malignant, or both type of tumors, the category of which covers carcinoma, fibroadenoma, cyst and abscess. The ground truths of tumors are labeled by two radiologists with 3 years of experience on breast MRI reporting by drawing contours of tumor slice by slice. ADUNet is compared with other prevalent deep-learning methods on the tumor segmentation and quantitative results, and achieves the best performance on both Case Dice Score and Global Dice Score by 0.748 and 0.801 respectively. Moreover, the customized classification network outperforms two CNN-M based models and achieves tumor-level and case-level AUC by 0.831 and 0.918 respectively.<\/jats:p><\/jats:sec><jats:sec><jats:title>Discussion<\/jats:title><jats:p>All data in this paper are collected from the same MRI device, thus it is reasonable to assume that they are from the same domain and independent identically distributed. Whether the proposed algorithm is robust enough in a multi-source case still remains an open question. Each stage of the proposed algorithm is trained separately, which makes each stage more robust and converge faster. Such training strategy considers each stage as a separate task and does not take into account the relationships between tasks.<\/jats:p><\/jats:sec>","DOI":"10.3389\/fcomp.2022.1054158","type":"journal-article","created":{"date-parts":[[2022,11,24]],"date-time":"2022-11-24T20:27:04Z","timestamp":1669321624000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["A coarse-to-fine and automatic algorithm for breast diagnosis on multi-series MRI images"],"prefix":"10.3389","volume":"4","author":[{"given":"Hongwei","family":"Xue","sequence":"first","affiliation":[]},{"given":"Guangwu","family":"Qian","sequence":"additional","affiliation":[]},{"given":"Xiaofen","family":"Wu","sequence":"additional","affiliation":[]},{"given":"Yan","family":"Gao","sequence":"additional","affiliation":[]},{"given":"Hongguang","family":"Yang","sequence":"additional","affiliation":[]},{"given":"Mingqian","family":"Liu","sequence":"additional","affiliation":[]},{"given":"Lei","family":"Wang","sequence":"additional","affiliation":[]},{"given":"Renfei","family":"Chen","sequence":"additional","affiliation":[]},{"given":"Peijun","family":"Wang","sequence":"additional","affiliation":[]}],"member":"1965","published-online":{"date-parts":[[2022,11,24]]},"reference":[{"key":"B1","article-title":"\u201cClassification of breast mri lesions using small-size training sets: comparison of deep learning approaches,\u201d","volume-title":"SPIE Medical Imaging","author":"Amit","year":"2017"},{"key":"B2","doi-asserted-by":"publisher","first-page":"637","DOI":"10.1109\/TMI.2012.2219589","article-title":"A multichannel markov random field framework for tumor segmentation with an application to classification of gene expression-based breast cancer recurrence risk","volume":"32","author":"Ashraf","year":"2013","journal-title":"IEEE Trans. Med. Imaging"},{"key":"B3","doi-asserted-by":"publisher","first-page":"680","DOI":"10.1148\/radiol.09090838","article-title":"Cancerous breast lesions on dynamic contrast-enhanced mr images: computerized characterization for image-based prognostic markers","volume":"254","author":"Bhooshan","year":"2010","journal-title":"Radiology"},{"key":"B4","doi-asserted-by":"publisher","first-page":"394","DOI":"10.3322\/caac.21492","article-title":"Global cancer statistics 2018: Globocan estimates of incidence and mortality worldwide for 36 cancers in 185 countries","volume":"68","author":"Bray","year":"2018","journal-title":"CA"},{"key":"B5","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR.2019.00511","article-title":"\u201cHybrid task cascade for instance segmentation,\u201d","volume-title":"IEEE Conference on Computer Vision and Pattern Recognition","author":"Chen","year":"2019"},{"key":"B6","doi-asserted-by":"publisher","first-page":"63","DOI":"10.1016\/j.acra.2005.08.035","article-title":"A fuzzy c-means (fcm)-based approach for computerized segmentation of breast lesions in dynamic contrast-enhanced mr images1","volume":"13","author":"Chen","year":"2006","journal-title":"Acad Radiol"},{"key":"B7","unstructured":"CholletF.\n          Keras. GitHub2015"},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1702.05970","article-title":"Automatic liver and tumor segmentation of ct and mri volumes using cascaded fully convolutional neural networks","author":"Christ","year":"2017","journal-title":"arXiv preprint arXiv:1702.05970"},{"key":"B9","first-page":"424","article-title":"\u201c3D u-net: learning dense volumetric segmentation from sparse annotation,\u201d","volume-title":"International Conference on Medical Image Computing and Computer-Assisted Intervention","author":"\u00c7i\u00e7ek","year":"2016"},{"key":"B10","doi-asserted-by":"crossref","first-page":"248","DOI":"10.1109\/CVPR.2009.5206848","article-title":"\u201cImagenet: a large-scale hierarchical image database,\u201d","volume-title":"2009 IEEE Conference on Computer Vision and Pattern Recognition","author":"Deng","year":"2009"},{"key":"B11","first-page":"147","article-title":"Breast tumor segmentation and classification using svm and bayesian from thermogram images","volume":"2","author":"Dinsha","year":"2014","journal-title":"Unique J. Eng. Adv. Sci"},{"key":"B12","doi-asserted-by":"crossref","DOI":"10.1590\/0100-3984.2015.0124","article-title":"\u201cRole of magnetic resonance imaging in the planning of breast cancer treatment strategies: comparison with conventional imaging techniques,\u201d","author":"Frana","year":"2017","journal-title":"Radiologia Brasileira"},{"key":"B13","doi-asserted-by":"publisher","first-page":"823","DOI":"10.1002\/mrm.1910030602","article-title":"Rare imaging: a fast imaging method for clinical mr","volume":"3","author":"Hennig","year":"1986","journal-title":"Mag. Reson. Med"},{"key":"B14","doi-asserted-by":"publisher","first-page":"219","DOI":"10.1016\/j.diii.2019.02.008","article-title":"Detection and characterization of mri breast lesions using deep learning","volume":"100","author":"Herent","year":"2019","journal-title":"Diagn. Interv. Imaging"},{"key":"B15","first-page":"2261","article-title":"\u201cDensely connected convolutional networks,\u201d","volume-title":"IEEE Conference on Computer Vision and Pattern Recognition","author":"Huang","year":"2017"},{"key":"B16","first-page":"448","article-title":"\u201cBatch normalization: accelerating deep network training by reducing internal covariate shift,\u201d","volume-title":"International Conference on Machine Learning","author":"Ioffe","year":"2015"},{"key":"B17","first-page":"1587","article-title":"\u201cLearning-based automatic breast tumor detection and segmentation in ultrasound images,\u201d","volume-title":"IEEE International Symposium on Biomedical Imaging","author":"Jiang","year":"2012"},{"key":"B18","doi-asserted-by":"publisher","first-page":"1921","DOI":"10.1148\/rg.2018180056","article-title":"American joint committee on cancer's staging system for breast cancer, eighth edition: what the radiologist needs to know","volume":"38","author":"Kalli","year":"2018","journal-title":"Radiographics"},{"key":"B19","doi-asserted-by":"publisher","first-page":"21","DOI":"10.1016\/j.media.2018.10.004","article-title":"Fully convolutional multi-scale residual densenets for cardiac segmentation and automated cardiac diagnosis using ensemble of classifiers","volume":"51","author":"Khened","year":"2019","journal-title":"Med. Image Anal"},{"key":"B20","unstructured":"Adam: A method for stochastic optimization\n            KingmaD. P.\n            Ba.J\n          arXiv [Preprint]2015"},{"key":"B21","article-title":"\u201cPyramid attention network for semantic segmentation,\u201d","volume-title":"British Machine Vision Conference","author":"Li","year":""},{"key":"B22","doi-asserted-by":"publisher","first-page":"2663","DOI":"10.1109\/TMI.2018.2845918","article-title":"H-denseunet: hybrid densely connected unet for liver and tumor segmentation from ct volumes","volume":"37","author":"Li","year":"","journal-title":"IEEE Trans. Med. Imaging"},{"key":"B23","first-page":"2980","article-title":"\u201cFocal loss for dense object detection,\u201d","volume-title":"Proceedings of the IEEE International Conference on Computer Vision","author":"Lin","year":"2017"},{"key":"B24","first-page":"3431","article-title":"\u201cFully convolutional networks for semantic segmentation,\u201d","volume-title":"IEEE Conference on Computer Vision and Pattern Recognition","author":"Long","year":"2015"},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-32251-9_55","article-title":"Deep angular embedding and feature correlation attention for breast mri cancer analysis","author":"Luo","year":"2019","journal-title":"arXiv preprint arXiv:1906.02999"},{"key":"B26","first-page":"546","article-title":"\u201cTraining medical image analysis systems like radiologists,\u201d","volume-title":"International Conference on Medical Image Computing and Computer-Assisted Intervention","author":"Maicas","year":"2018"},{"key":"B27","first-page":"565","article-title":"\u201cV-net: fully convolutional neural networks for volumetric medical image segmentation,\u201d","volume-title":"International Conference on 3D Vision","author":"Milletari","year":"2016"},{"key":"B28","first-page":"2204","article-title":"\u201cRecurrent models of visual attention,\u201d","volume-title":"Advances in Neural Information Processing Systems","author":"Mnih","year":"2014"},{"key":"B29","article-title":"\u201cAttention u-net: learning where to look for the pancreas,\u201d","volume-title":"Medical Imaging with Deep Learning","author":"Oktay","year":"2018"},{"key":"B30","doi-asserted-by":"publisher","first-page":"1007","DOI":"10.1049\/iet-cvi.2018.5273","article-title":"Comprehensive computer-aided diagnosis for breast t1-weighted dce-mri through quantitative dynamical features and spatio-temporal local binary patterns","volume":"12","author":"Piantadosi","year":"2018","journal-title":"IET Comput. Vis"},{"key":"B31","article-title":"Fusionnet: a deep fully residual convolutional neural network for image segmentation in connectomics","author":"Quan","year":"2016","journal-title":"arXiv preprint arXiv:1612.05360"},{"key":"B32","first-page":"234","article-title":"\u201cU-net: convolutional networks for biomedical image segmentation,\u201d","volume-title":"International Conference on Medical Image Computing and Computer-Assisted Intervention","author":"Ronneberger","year":"2015"},{"key":"B33","doi-asserted-by":"publisher","first-page":"94","DOI":"10.1016\/j.media.2018.01.006","article-title":"Spatial aggregation of holistically-nested convolutional neural networks for automated pancreas localization and segmentation","volume":"45","author":"Roth","year":"2017","journal-title":"Med. Image Anal"},{"key":"B34","doi-asserted-by":"publisher","first-page":"e0190348","DOI":"10.1371\/journal.pone.0190348","article-title":"Automatic outer and inner breast tissue segmentation using multi-parametric mri images of breast tumor patients","volume":"13","author":"Thakran","year":"2018","journal-title":"PLoS ONE"},{"key":"B35","doi-asserted-by":"publisher","first-page":"444","DOI":"10.1158\/1055-9965.EPI-16-0858","article-title":"Global cancer in women: burden and trends","volume":"26","author":"Torre","year":"2017","journal-title":"Cancer Epidemiol. Biomark. Prevent"},{"key":"B36","doi-asserted-by":"publisher","first-page":"1341","DOI":"10.1002\/jmri.22680","article-title":"Performance of a fully automatic lesion detection system for breast dce-mri","volume":"34","author":"Vignati","year":"2011","journal-title":"J. Mag. Reson. Imaging"},{"key":"B37","first-page":"6450","article-title":"\u201cResidual attention network for image classification,\u201d","volume-title":"Computer Vision and Pattern Recognition","author":"Wang","year":"2017"},{"key":"B38","first-page":"321","article-title":"\u201cAttention-aware path-based relation extraction for medical knowledge graph,\u201d","volume-title":"International Conference on Smart Computing and Communication","author":"Wen","year":"2017"},{"key":"B39","doi-asserted-by":"publisher","first-page":"742","DOI":"10.1007\/s11263-019-01198-w","article-title":"Group normalization","volume":"128","author":"Wu","year":"2020","journal-title":"Int. J. Comput. Vis"},{"key":"B40","doi-asserted-by":"publisher","first-page":"94","DOI":"10.1109\/JSTSP.2008.2011110","article-title":"Breast tumor analysis in dynamic contrast enhanced mri using texture features and wavelet transform","volume":"3","author":"Yao","year":"2009","journal-title":"IEEE J. Sel. Top. Signal Process"},{"key":"B41","doi-asserted-by":"crossref","DOI":"10.18653\/v1\/W16-0103","article-title":"\u201cAttention-based convolutional neural network for machine comprehension,\u201d","volume-title":"Proceedings of the Workshop on Human-Computer Question Answering. Association for Computational Linguistics","author":"Yin","year":"2016"},{"key":"B42","first-page":"66","article-title":"\u201cVolumetric convnets with mixed residual connections for automated prostate segmentation from 3D MR images,\u201d","volume-title":"AAAI","author":"Yu","year":"2017"},{"key":"B43","doi-asserted-by":"publisher","first-page":"1876","DOI":"10.1109\/TMI.2017.2695227","article-title":"Automatic skin lesion segmentation using deep fully convolutional networks with jaccard distance","volume":"36","author":"Yuan","year":"2017","journal-title":"IEEE Trans. Med. Imaging"},{"key":"B44","doi-asserted-by":"publisher","first-page":"1158","DOI":"10.1016\/j.acra.2010.04.015","article-title":"Multimodality computer-aided breast cancer diagnosis with ffdm and DCE-MRI","volume":"17","author":"Yuan","year":"2010","journal-title":"Acad. Radiol"},{"key":"B45","first-page":"720","article-title":"\u201cJoint craniomaxillofacial bone segmentation and landmark digitization by context-guided fully convolutional networks,\u201d","volume-title":"International Conference on Medical Image Computing and Computer-Assisted Intervention","author":"Zhang","year":"2017"},{"key":"B46","first-page":"393","article-title":"\u201cSegmentation and classification of breast tumor using dynamic contrast-enhanced mr images,\u201d","volume-title":"International Conference on Medical Image Computing and Computer-Assisted Intervention","author":"Zheng","year":"2007"},{"key":"B47","doi-asserted-by":"publisher","first-page":"1144","DOI":"10.1002\/jmri.26721","article-title":"Weakly supervised 3D deep learning for breast cancer classification and localization of the lesions in mr images","volume":"50","author":"Zhou","year":"2019","journal-title":"J. Mag. Reson. Imaging"}],"container-title":["Frontiers in Computer Science"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fcomp.2022.1054158\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,11,24]],"date-time":"2022-11-24T20:27:25Z","timestamp":1669321645000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fcomp.2022.1054158\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,11,24]]},"references-count":47,"alternative-id":["10.3389\/fcomp.2022.1054158"],"URL":"https:\/\/doi.org\/10.3389\/fcomp.2022.1054158","relation":{},"ISSN":["2624-9898"],"issn-type":[{"type":"electronic","value":"2624-9898"}],"subject":[],"published":{"date-parts":[[2022,11,24]]},"article-number":"1054158"}}