Abstract To accurately identify the microcalcification in X-ray images for diagnosis and early prevention of breast cancer, a microcalcification object detection method combining fine-grained cascading enhancement-network(FCE-Net) and multi-scale feature fusion(MFF]is proposed. Firstly, the residual weights of FCE-Net are accumulated and the multi-branch structure is enhanced within the convolution module to obtain hierarchical and fine-grained convolution plots. Then, a candidate detection network based on MFF is constructed and multi-scale features are merged by double-upsampling. Thus, the confidence and regional coordinates of the object of the microcalcification clusters are acquired. Finally, the object region is classified and the bounding box is adjusted in the pooling layer of the region of interest. The experimental results on MIAS breast cancer dataset show that the method combining FCE-Net and MFF has a better ability to extract deep feature with enhanced classification and positioning accuracies.
Fund:Supported by General Program of National Natural Science Foun-dation of China(No.41877527), Natural Science Foundation of Shaanxi Province(No.2016JM6023)
Corresponding Authors:
ZHANG Xinsheng,Ph.D., professor. His research interests include pattern recognition and intelligent information processing.
About author:: WANG Zhe, master student. His research interests include pattern recognition and inte-lligent]information processing.
ZHANG Xinsheng,WANG Zhe. Microcalcification Clusters Detection Method Based on Deep Learning and Multi-scale Feature Fusion[J]. , 2018, 31(11): 1028-1039.
[1]封任冬,汪华,丁莹莹,等.乳腺癌筛查模式的探讨.临床放射学杂志, 2016, 35(1): 36-40. (FENG R D, WANG H, DING Y Y, et al.To Investigate the Screening Model of Breast Cancer. Journal of Clinical Radiology, 2016, 35(1): 36-40.) [2]JEH S K, KIM S H, CHOI J J, et al. Comparison of Automated Breast Ultrasonography to Handheld Ultrasonography in Detecting and Diagnosing Breast Lesions. Acta Radiologica, 2016, 57(2): 162-169. [3]陈万青,郑荣寿.中国女性乳腺癌发病死亡和生存状况.中国肿瘤临床, 2015, 42(13): 668-674. (CHEN W Q, ZHENG R S. Incidence, Mortality and Survival Analysis]of Breast Cancer in China. Chinese Journal of Clinical Oncology, 2015, 42(13): 668-674.) [4]王磊,朱淼良,邓丽萍,等.一种基于二维粒子的自动检测乳腺钼靶片上微钙化点簇的方法.计算机研究与发展, 2009, 46(9):1438-1445. (WANG L, ZHU M L, DENG L P, et al. Automatic Detection on Clustered Microcalcifications on Mammograms Based on 2D Parti-cles. Journal of Computer Research and Development, 2009, 46(9):1438-1445.) [5]汪登斌,李志,王丽君,等.MRI对乳腺X线摄影中含成簇微钙化病变的诊断.放射学实践, 2012, 27(10): 1089-1094. (WANG D B, LI Z, WANG L J, et al. MRI in the Diagnosis of Breast Lesions with Clustered Micro-calcifications Showing on Mammography. Radio Practice, 2012, 27(10): 1089-1094.) [6]曹鹏,李博,刘鑫,等.基于代价敏感SVM优化组合算法的微钙化簇识别.东北大学学报(自然科学版), 2013, 34(8): 1100-1104. (CAO P, LI B, LIU X, et al. Microcalcification Clusters Recognition Based on Optimized Cost-Sensitive SVM Combinational Algorithm. Journal of Northeastern University(Natural Science), 2013, 34(8): 1100-1104.) [7]KOOI T, LITJENS G, VAN GINNEKEN B, et al. Large Scale Deep Learning for Computer Aided Detection of Mammographic Lesions. Medical Image Analysis, 2017, 35: 303-312. [8]WANG K J, DONG M, YANG Z, et al. Regions of Micro-calcifications Clusters Detection Based on New Features from Imbalance Data in Mammograms // Proc of the 8th International Conference on Graphic and Image Processing. Berlin, Germany: Springer, 2016, 272-282. [9]王红强,顾康生.肿瘤精准医疗中的个性化药效评估(预测)方法.模式识别与人工智能, 2017, 30(2): 117-126. (WANG H Q, GU K S. Methods for Personalized Drug Effectiveness Prediction in Cancer Precision Medicine. Pattern Recognition and Artificial Intelligence, 2017, 30(2): 117-126. [10]CAO C, HAYWARD R B, MÜLLER M. Move Prediction Using Deep Convolutional Neural Networks in Hex[C/OL]. [2018-04-25]. https://webdocs.cs.ualberta.ca/~hayward/papers/movepredhex.pdf. [11]GIRSHICK R. Fast R-CNN // Proc of the IEEE International Conference on Computer Vision. Washington, USA: IEEE, 2015: 1440-1448. [12]HE K M, ZHANG X Y, REN S Q, et al. Deep Residual Learning for Image Recognition // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2016: 770-778. [13]ZAGORUYKO S, KOMODAKIS N. Wide Residual Networks[C/OL]. [2018-04-25]. https://arxiv.org/pdf/1605.07146.pdf. [14]WU Z F, SHEN C H, VAN DEN HENGEL A. Wider or Deeper: Revisiting the ResNet Model for Visual Recognition[C/OL]. [2018-04-25]. https://arxiv.org/pdf/1611.10080.pdf. [15]LEE Y, YIM B, KIM H, et al. Wide-Residual-Inception Networks for Real-Time Object Detection[J/OL]. [2018-04-25]. https://arxiv.org/ftp/arxiv/papers/1702/1702.01243.pdf. [16]SZEGEDY C, VANHOUCKE V, IOFFE S, et al. Rethinking the Inception Architecture for Computer Vision // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2016: 2818-2826. [17]SZEGEDY C, IOFFE S, VANHOUCKE V, et al. Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning[C/OL]. [2018-04-25]. https://arxiv.org/pdf/1602.07261.pdf. [18]REDMON J, DIVVALA S, GIRSHICK R, et al. You Only Look Once: Unified, Real-Time Object Detection[C/OL]. [2018-04-25]. https://pjreddie.com/media/files/papers/yolo.pdf. [19]REN S Q, HE K M, GIRSHICK R, et al. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6): 1137-1149. [20]HE K M, ZHANG X Y, REN S Q, et al. Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2014, 37(9): 1904-1916. [21]LIN T Y, DOLLAR P, GIRSHICK R, et al.]Feature Pyramid Networks for Object Detection // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2016: 936-944. [22]YIP M, MACKENZIE A, LEWIS E, et al. Image Resampling Effects in Mammographic Image Simulation. Physics in Medicine and Biology, 2011, 56(22): 275-286. [23]LI W S, DONG P, XIAO B, et al. Object Recognition Based on the Region of Interest and Optimal Bag of Words Model. Neurocomputing, 2013, 172: 394-398. [24]KANAMORI T. Deformation of Log-Likelihood Loss Function for Multiclass Boosting. Neural Networks, 2010, 23(7): 843-864. [25]HE K M, ZHANG X Y, REN S Q, et al. Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification // Proc of the IEEE International Conference on Computer Vision. Washington, USA: IEEE, 2015: 1026-1034. ][26]NESTEROV Y, SPOKOINY V. Random Gradient-Free Minimization of Convex Functions. Foundations of Computational Mathema-tics, 2017, 17(2): 527-566. [27]BOTEV A, LEVER G, BARBER D. Nesterov's Accelerated Gradient and Momentum as Approximations to Regularised Update Descent // Proc of the International Joint Conference on Neural Networks. New York, USA: ACM, 2017: 1899-1903. [28]LIU W Y, WEN Y D, YU Z D, et al. Large-Margin Softmax Loss for Convolutional Neural Networks // Proc of the 33rd International Conference on Machine Learning. New York, USA: ACM, 2016: 507-516. [29]KLINE D M, BERARDI V L. Revisiting Squared-Error and Cross-Entropy Functions for Training Neural Network Classifiers. Neural Computing and Applications, 2005, 14(4): 310-318. [30]ZHANG T Z, CAI R, LI Z W, et al. Hierarchical Object Representations for Visual Recognition via Weakly Supervised Learning // Proc of the Asian Conference on Computer Vision. Berlin, Germany: Springer, 2012: 474-485. [31]吴枰,潘海为,高琳琳,等.基于KAP有向图模型的医学图像分类算法.模式识别与人工智能, 2016, 29(5): 427-438. (WU P, PAN H W, GAO L L, et al.Medical Image Classification Algorithm Based on KAP Directed Graph Model. Pattern Recogni-tion and Artificial Intelligence, 2016, 29(5): 427-438.)
2018, Vol. 31 
