Deep Neural Networks for Breast Cancer Diagnosis: Fine Needle Biopsy Scenario

In this study, we focus on the problem of computer-aided diagnosis of breast cancer using cytological images of fine needle biopsies. We explore the potential of modern deep neural network architectures by comparing five different convolutional neural networks trained to classify the specimen as either benign or malignant. For experimentation, we use 550 cytological images of fine needle biopsies from 50 patients, balanced between benign and malignant cases, acquired at the University Hospital in Zielona Gora, Poland. We found that the convolutional neural network Inception-v3 is the best model, reaching 91.86% accuracy and 0.97 value for area under the curve (AUC).

[1]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[2]  Marek Kowal,et al.  Classification of breast cancer cytological specimen using convolutional neural network , 2017 .

[3]  Marek Kowal,et al.  Computer-aided diagnosis of breast cancer based on fine needle biopsy microscopic images , 2013, Comput. Biol. Medicine.

[4]  Dumitru Erhan,et al.  Going deeper with convolutions , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[5]  Michel E. Vandenberghe,et al.  Relevance of deep learning to facilitate the diagnosis of HER2 status in breast cancer , 2017, Scientific Reports.

[6]  P. D. Britton,et al.  Fine needle aspiration or core biopsy , 1999 .

[7]  James Cressee Elphinstone Underwood.  INTRODUCTION TO BIOPSY INTERPRETATION AND SURGICAL PATHOLOGY , 1981 .

[8]  Vijay Vasudevan,et al.  Learning Transferable Architectures for Scalable Image Recognition , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[9]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

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

[11]  Li Fei-Fei,et al.  ImageNet: A large-scale hierarchical image database , 2009, CVPR.

[12]  Kaiming He,et al.  Focal Loss for Dense Object Detection , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[13]  S. Ciatto,et al.  Accuracy of needle biopsy of breast lesions visible on ultrasound: audit of fine needle versus core needle biopsy in 3233 consecutive samplings with ascertained outcomes. , 2012, Breast.

[14]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[15]  Marek Kowal,et al.  Nuclei segmentation for computer-aided diagnosis of breast cancer , 2014, Int. J. Appl. Math. Comput. Sci..

[16]  Marek Kowal,et al.  Automatic Breast Cancer Diagnosis Based on K-Means Clustering and Adaptive Thresholding Hybrid Segmentation , 2011, IP&C.

[17]  J. Weydert,et al.  Fine Needle Aspiration: Current Practice and Recent Developments , 2003 .

[18]  Yoshua Bengio,et al.  Understanding the difficulty of training deep feedforward neural networks , 2010, AISTATS.

[19]  Yasmeen M. George,et al.  Automated cell nuclei segmentation for breast fine needle aspiration cytology , 2013, Signal Process..

[20]  Sebastian Thrun,et al.  Dermatologist-level classification of skin cancer with deep neural networks , 2017, Nature.

[21]  Kilian Q. Weinberger,et al.  Densely Connected Convolutional Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[22]  Sergey Ioffe,et al.  Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning , 2016, AAAI.

[23]  Xin Sun,et al.  Evaluation and accurate diagnoses of pediatric diseases using artificial intelligence , 2019, Nature Medicine.

[24]  A. Jemal,et al.  Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries , 2018, CA: a cancer journal for clinicians.

[25]  D Parmar,et al.  Diagnostic application of computerised nuclear morphometric image analysis in fine needle aspirates of breast lesions , 2015 .

[26]  Forrest N. Iandola,et al.  SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <1MB model size , 2016, ArXiv.

[27]  Roman Monczak,et al.  Computer-Aided Breast Cancer Diagnosis Based on the Analysis of Cytological Images of Fine Needle Biopsies , 2013, IEEE Transactions on Medical Imaging.