Deep-learning approach with convolutional neural network for classification of maximum intensity projections of dynamic contrast-enhanced breast magnetic resonance imaging.

PURPOSE We aimed to evaluate deep learning approach with convolutional neural networks (CNNs) to discriminate between benign and malignant lesions on maximum intensity projections of dynamic contrast-enhanced breast magnetic resonance imaging (MRI). METHODS We retrospectively gathered maximum intensity projections of dynamic contrast-enhanced breast MRI of 106 benign (including 22 normal) and 180 malignant cases for training and validation data. CNN models were constructed to calculate the probability of malignancy using CNN architectures (DenseNet121, DenseNet169, InceptionResNetV2, InceptionV3, NasNetMobile, and Xception) with 500 epochs and analyzed that of 25 benign (including 12 normal) and 47 malignant cases for test data. Two human readers also interpreted these test data and scored the probability of malignancy for each case using Breast Imaging Reporting and Data System. Sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUC) were calculated. RESULTS The CNN models showed a mean AUC of 0.830 (range, 0.750-0.895). The best model was InceptionResNetV2. This model, Reader 1, and Reader 2 had sensitivities of 74.5%, 72.3%, and 78.7%; specificities of 96.0%, 88.0%, and 80.0%; and AUCs of 0.895, 0.823, and 0.849, respectively. No significant difference arose between the CNN models and human readers (p > 0.125). CONCLUSION Our CNN models showed comparable diagnostic performance in differentiating between benign and malignant lesions to human readers on maximum intensity projection of dynamic contrast-enhanced breast MRI.

[1]  Ukihide Tateishi,et al.  Detection and Diagnosis of Breast Cancer Using Artificial Intelligence Based Assessment of Maximum Intensity Projection Dynamic Contrast-Enhanced Magnetic Resonance Images , 2020, Diagnostics.

[2]  Sergey Ioffe,et al.  Rethinking the Inception Architecture for Computer Vision , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[3]  E. Mendelson,et al.  Artificial Intelligence in Breast Imaging: Potentials and Limitations. , 2019, AJR. American journal of roentgenology.

[4]  E. Fukuma,et al.  Feasibility and potential limitations of abbreviated breast MRI: an observer study using an enriched cohort , 2017, Breast Cancer.

[5]  P. Hérent,et al.  Detection and characterization of MRI breast lesions using deep learning. , 2019, Diagnostic and interventional imaging.

[6]  Pheng-Ann Heng,et al.  Weakly supervised 3D deep learning for breast cancer classification and localization of the lesions in MR images , 2019, Journal of magnetic resonance imaging : JMRI.

[7]  Ralf-Dieter Hilgers,et al.  Abbreviated breast magnetic resonance imaging (MRI): first postcontrast subtracted images and maximum-intensity projection-a novel approach to breast cancer screening with MRI. , 2014, Journal of clinical oncology : official journal of the American Society of Clinical Oncology.

[8]  Ukihide Tateishi,et al.  Breast Ultrasound Image Synthesis using Deep Convolutional Generative Adversarial Networks , 2019, Diagnostics.

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

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

[11]  F J Gilbert,et al.  Artificial intelligence in breast imaging. , 2019, Clinical radiology.

[12]  G. Newstead,et al.  Decision making for breast lesions initially detected at contrast-enhanced breast MRI. , 2013, AJR. American journal of roentgenology.

[13]  Y. Kanda,et al.  Investigation of the freely available easy-to-use software ‘EZR' for medical statistics , 2012, Bone Marrow Transplantation.

[14]  Paul Babyn,et al.  Generative Adversarial Network in Medical Imaging: A Review , 2018, Medical Image Anal..

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

[16]  M. Yaffe,et al.  American Cancer Society Guidelines for Breast Screening with MRI as an Adjunct to Mammography , 2007 .

[17]  Dorit Merhof,et al.  Radiomic versus Convolutional Neural Networks Analysis for Classification of Contrast-enhancing Lesions at Multiparametric Breast MRI. , 2019, Radiology.

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

[19]  Deepa Sheth,et al.  Artificial intelligence in the interpretation of breast cancer on MRI , 2019, Journal of magnetic resonance imaging : JMRI.

[20]  Ernst J. Rummeny,et al.  Intra- and inter-observer variability in measurement of target lesions: implication on response evaluation according to RECIST 1.1 , 2012, Radiology and oncology.

[21]  T. M. Kolb,et al.  Comparison of the performance of screening mammography, physical examination, and breast US and evaluation of factors that influence them: an analysis of 27,825 patient evaluations. , 2002, Radiology.

[22]  P. Langenberg,et al.  Breast Imaging Reporting and Data System: inter- and intraobserver variability in feature analysis and final assessment. , 2000, AJR. American journal of roentgenology.

[23]  Cheng Li,et al.  A Radiomics Approach With CNN for Shear-Wave Elastography Breast Tumor Classification , 2018, IEEE Transactions on Biomedical Engineering.

[24]  Hui Sun,et al.  AUNet: attention-guided dense-upsampling networks for breast mass segmentation in whole mammograms , 2018, Physics in medicine and biology.

[25]  Hairong Zheng,et al.  Multi-View Mammographic Density Classification by Dilated and Attention-Guided Residual Learning , 2020, IEEE/ACM Transactions on Computational Biology and Bioinformatics.

[26]  Karel G M Moons,et al.  Meta-analysis of MR imaging in the diagnosis of breast lesions. , 2008, Radiology.

[27]  M. Tillich,et al.  Breast MRI used as a problem-solving tool reliably excludes malignancy. , 2015, European journal of radiology.

[28]  Hiroyuki Abe,et al.  Use of clinical MRI maximum intensity projections for improved breast lesion classification with deep convolutional neural networks , 2018, Journal of medical imaging.

[29]  Ukihide Tateishi,et al.  Distinction between benign and malignant breast masses at breast ultrasound using deep learning method with convolutional neural network , 2019, Japanese Journal of Radiology.