Deep learning-based segmentation of breast masses in dedicated breast CT imaging: Radiomic feature stability between radiologists and artificial intelligence

A deep learning (DL) network for 2D-based breast mass segmentation in unenhanced dedicated breast CT images was developed and validated, and its robustness in radiomic feature stability and diagnostic performance compared to manual annotations of multiple radiologists was investigated. 93 mass-like lesions were extensively augmented and used to train the network (n = 58 masses), which was then tested (n = 35 masses) against manual ground truth of a qualified breast radiologist with experience in breast CT imaging using the Conformity coefficient (with a value equal to 1 indicating a perfect performance). Stability and diagnostic power of 672 radiomic descriptors were investigated between the computerized segmentation, and 4 radiologists' annotations for the 35 test set cases. Feature stability and diagnostic performance in the discrimination between benign and malignant cases were quantified using intraclass correlation (ICC) and multivariate analysis of variance (MANOVA), performed for each segmentation case (4 radiologists and DL algorithm). DL-based segmentation resulted in a Conformity of 0.85 ± 0.06 against the annotated ground truth. For the stability analysis, although modest agreement was found among the four annotations performed by radiologists (Conformity 0.78 ± 0.03), over 90% of all radiomic features were found to be stable (ICC>0.75) across multiple segmentations. All MANOVA analyses were statistically significant (p ≤ 0.05), with all dimensions equal to 1, and Wilks' lambda ≤0.35. In conclusion, DL-based mass segmentation in dedicated breast CT images can achieve high segmentation performance, and demonstrated to provide stable radiomic descriptors with comparable discriminative power in the classification of benign and malignant tumors to expert radiologist annotation.

[1]  Peter Bajcsy,et al.  Cell Image Segmentation Using Generative Adversarial Networks, Transfer Learning, and Augmentations , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[2]  Francesca Caumo,et al.  An exploratory radiomics analysis on digital breast tomosynthesis in women with mammographically negative dense breasts. , 2018, Breast.

[3]  J. Fleiss,et al.  Intraclass correlations: uses in assessing rater reliability. , 1979, Psychological bulletin.

[4]  Ritse Mann,et al.  Automated soft tissue lesion detection and segmentation in digital mammography using a u-net deep learning network , 2018, ArXiv.

[5]  Tae-Seong Kim,et al.  A fully integrated computer-aided diagnosis system for digital X-ray mammograms via deep learning detection, segmentation, and classification , 2018, Int. J. Medical Informatics.

[6]  Karen Drukker,et al.  Combined Benefit of Quantitative Three-Compartment Breast Image Analysis and Mammography Radiomics in the Classification of Breast Masses in a Clinical Data Set. , 2019, Radiology.

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

[8]  Anne L. Martel,et al.  Semi-Automatic Region-of-Interest Segmentation Based Computer-Aided Diagnosis of Mass Lesions from Dynamic Contrast-Enhanced Magnetic Resonance Imaging Based Breast Cancer Screening , 2014, Journal of Digital Imaging.

[9]  Terry K Koo,et al.  A Guideline of Selecting and Reporting Intraclass Correlation Coefficients for Reliability Research. , 2016, Journal Chiropractic Medicine.

[10]  Masoom A. Haider,et al.  Discovery Radiomics for Pathologically-Proven Computed Tomography Lung Cancer Prediction , 2015, ICIAR.

[11]  Albert Gubern-Mérida,et al.  Automated lesion detection and segmentation in digital mammography using a u-net deep learning network , 2018, Other Conferences.

[12]  Bram van Ginneken,et al.  Pulmonary Nodule Detection in CT Images: False Positive Reduction Using Multi-View Convolutional Networks , 2016, IEEE Transactions on Medical Imaging.

[13]  Jie Tian,et al.  Multiparametric MRI-based radiomics analysis for prediction of breast cancers insensitive to neoadjuvant chemotherapy , 2019, Clinical and Translational Oncology.

[14]  Ioannis Sechopoulos,et al.  Breast parenchyma analysis and classification for breast masses detection using texture feature descriptors and neural networks in dedicated breast CT images , 2019, Medical Imaging.

[15]  Olivier Gevaert,et al.  Core samples for radiomics features that are insensitive to tumor segmentation: method and pilot study using CT images of hepatocellular carcinoma , 2015, Journal of medical imaging.

[16]  Mohamed Cheriet,et al.  A new framework for online sketch-based image retrieval in web environment , 2012, 2012 11th International Conference on Information Science, Signal Processing and their Applications (ISSPA).

[17]  Russell T Warne.  A primer on multivariate analysis of variance (MANOVA) for behavioral scientists , 2014 .

[18]  P. Lambin,et al.  Robust Radiomics Feature Quantification Using Semiautomatic Volumetric Segmentation , 2014, PloS one.

[19]  Kenneth I. Laws,et al.  Rapid Texture Identification , 1980, Optics & Photonics.

[20]  J. Boone,et al.  Dedicated breast CT: radiation dose and image quality evaluation. , 2001, Radiology.

[21]  Andriy Fedorov,et al.  Computational Radiomics System to Decode the Radiographic Phenotype. , 2017, Cancer research.

[22]  Soumith Chintala,et al.  Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks , 2015, ICLR.

[23]  Si Eun Lee,et al.  Radiomics of US texture features in differential diagnosis between triple-negative breast cancer and fibroadenoma , 2018, Scientific Reports.

[24]  Thomas Frauenfelder,et al.  Influence of inter-observer delineation variability on radiomics stability in different tumor sites , 2018, Acta oncologica.

[25]  Di Dong,et al.  2D and 3D CT Radiomics Features Prognostic Performance Comparison in Non-Small Cell Lung Cancer , 2017, Translational oncology.

[26]  Jie Ding,et al.  Task‐based assessment of a convolutional neural network for segmenting breast lesions for radiomic analysis , 2019, Magnetic resonance in medicine.

[27]  Patrick Granton,et al.  Radiomics: extracting more information from medical images using advanced feature analysis. , 2012, European journal of cancer.

[28]  Ioannis Sechopoulos,et al.  Dosimetric characterization of a dedicated breast computed tomography clinical prototype. , 2010, Medical physics.

[29]  Saman A. Zonouz,et al.  CloudID: Trustworthy cloud-based and cross-enterprise biometric identification , 2015, Expert Syst. Appl..

[30]  Maryellen L. Giger,et al.  Level Set Segmentation of Breast Masses in Contrast-Enhanced Dedicated Breast CT and Evaluation of Stopping Criteria , 2014, Journal of Digital Imaging.

[31]  Jianhua Liu,et al.  Image Feature Extraction Method Based on Shape Characteristics and Its Application in Medical Image Analysis , 2011, ICAIC.

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

[33]  Yuanjie Zheng,et al.  Parenchymal texture analysis in digital mammography: A fully automated pipeline for breast cancer risk assessment. , 2015, Medical physics.

[34]  Mandyam D. Srinath,et al.  Contour sequence moments for the classification of closed planar shapes , 1987, Pattern Recognit..

[35]  Matti Pietikäinen,et al.  Multiresolution Gray-Scale and Rotation Invariant Texture Classification with Local Binary Patterns , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[36]  Robert M. Haralick,et al.  Textural Features for Image Classification , 1973, IEEE Trans. Syst. Man Cybern..

[37]  Kevin P. Weinfurt.  Multivariate analysis of variance. , 1995 .

[38]  Mary M. Galloway,et al.  Texture analysis using gray level run lengths , 1974 .

[39]  James M. Keller,et al.  Texture description and segmentation through fractal geometry , 1989, Comput. Vis. Graph. Image Process..

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

[41]  Hui Li,et al.  Digital Mammography in Breast Cancer: Additive Value of Radiomics of Breast Parenchyma. , 2019, Radiology.

[42]  A. Vadivel,et al.  Effect of BIRADS shape descriptors on breast cancer analysis , 2015, Int. J. Medical Eng. Informatics.

[43]  Yuxing Tang,et al.  CT-realistic data augmentation using generative adversarial network for robust lymph node segmentation , 2019, Medical Imaging.

[44]  Ingrid Reiser,et al.  Optimal reconstruction and quantitative image features for computer‐aided diagnosis tools for breast CT , 2017, Medical physics.

[45]  Kpalma Kidiyo,et al.  A Survey of Shape Feature Extraction Techniques , 2008 .

[46]  P. Lambin,et al.  Stability of FDG-PET Radiomics features: An integrated analysis of test-retest and inter-observer variability , 2013, Acta oncologica.

[47]  Jan Hendrik Moltz,et al.  Stability of radiomic features of liver lesions from manual delineation in CT scans , 2019, Medical Imaging.

[48]  Viksit Kumar,et al.  Automated and real-time segmentation of suspicious breast masses using convolutional neural network , 2018, PloS one.

[49]  Gregory D. Hager,et al.  Adversarial deep structured nets for mass segmentation from mammograms , 2017, 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018).

[50]  Thomas Neff,et al.  Generative Adversarial Network based Synthesis for Supervised Medical Image Segmentation , 2017 .

[51]  Yoshua Bengio,et al.  Generative Adversarial Nets , 2014, NIPS.

[52]  Rangaraj M. Rangayyan,et al.  Application of shape analysis to mammographic calcifications , 1994, IEEE Trans. Medical Imaging.

[53]  Yuji Iwahori,et al.  Automatic Detection of Polyp Using Hessian Filter and HOG Features , 2015, KES.

[54]  Paul Kinahan,et al.  Radiomics: Images Are More than Pictures, They Are Data , 2015, Radiology.

[55]  J. Ferlay,et al.  Cancer incidence and mortality patterns in Europe: estimates for 40 countries in 2012. , 2013, European journal of cancer.

[56]  J. Boone,et al.  Dedicated breast CT: initial clinical experience. , 2008, Radiology.

[57]  Jinhua Yu,et al.  Automatic tumor segmentation in breast ultrasound images using a dilated fully convolutional network combined with an active contour model , 2018, Medical physics.