Breast cancer MRI radiomics: An overview of algorithmic features and impact of inter‐reader variability in annotating tumors

Purpose To review features used in MRI radiomics of breast cancer and study the inter‐reader stability of the features. Methods We implemented 529 algorithmic features that can be extracted from tumor and fibroglandular tissue (FGT) in breast MRIs. The features were identified based on a review of the existing literature with consideration of their usage, prognostic ability, and uniqueness. The set was then extended so that it comprehensively describes breast cancer imaging characteristics. The features were classified into 10 groups based on the type of data used to extract them and the type of calculation being performed. For the assessment of inter‐reader variability, four fellowship‐trained readers annotated tumors on preoperative dynamic contrast‐enhanced MRIs for 50 breast cancer patients. Based on the annotations, an algorithm automatically segmented the image and extracted all features resulting in one set of features for each reader. For a given feature, the inter‐reader stability was defined as the intraclass correlation coefficient (ICC) computed using the feature values obtained through all readers for all cases. Results The average inter‐reader stability for all features was 0.8474 (95% CI: 0.8068–0.8858). The mean inter‐reader stability was lower for tumor‐based features (0.6348, 95% CI: 0.5391–0.7257) than FGT‐based features (0.9984, 95% CI: 0.9970–0.9992). The feature group with the highest inter‐reader stability quantifies breast and FGT volume. The feature group with the lowest inter‐reader stability quantifies variations in tumor enhancement. Conclusions Breast MRI radiomics features widely vary in terms of their stability in the presence of inter‐reader variability. Appropriate measures need to be taken for reducing this variability in tumor‐based radiomics.

[1]  James C. Bezdek,et al.  Pattern Recognition with Fuzzy Objective Function Algorithms , 1981, Advanced Applications in Pattern Recognition.

[2]  Bonnie N Joe,et al.  Breast MR imaging for extent of disease assessment in patients with newly diagnosed breast cancer. , 2013, Magnetic resonance imaging clinics of North America.

[3]  R. Lerski,et al.  Interim heterogeneity changes measured using entropy texture features on T2-weighted MRI at 3.0 T are associated with pathological response to neoadjuvant chemotherapy in primary breast cancer , 2017, European Radiology.

[4]  Xiaozhi Yu,et al.  Effects of MRI scanner parameters on breast cancer radiomics , 2017, Expert Syst. Appl..

[5]  Woo Kyung Moon,et al.  Early Prediction of Response to Neoadjuvant Chemotherapy Using Parametric Response , 2015 .

[6]  M. R. Smith,et al.  Heterogeneity of Vascular Permeability in Breast Lesions with Dynamic Contrast Enhanced MRI , 2009 .

[7]  T. Uematsu,et al.  Background enhancement of mammary glandular tissue on breast dynamic MRI: imaging features and effect on assessment of breast cancer extent , 2012, Breast Cancer.

[8]  M. Giger,et al.  Cancerous breast lesions on dynamic contrast-enhanced MR images: computerized characterization for image-based prognostic markers. , 2010, Radiology.

[9]  M L Giger,et al.  Computerized analysis of breast lesions in three dimensions using dynamic magnetic-resonance imaging. , 1998, Medical physics.

[10]  Anant Madabhushi,et al.  A Radio-genomics Approach for Identifying High Risk Estrogen Receptor-positive Breast Cancers on DCE-MRI: Preliminary Results in Predicting OncotypeDX Risk Scores , 2016, Scientific Reports.

[11]  B. Yun,et al.  Quantitative analysis of breast parenchymal background enhancement (BPE) on magnetic resonance (MR) imaging: Association with mammographic breast density and aggressiveness of the primary cancer in postmenopausal women. , 2013, Journal of clinical oncology : official journal of the American Society of Clinical Oncology.

[12]  Ahmed Bilal Ashraf,et al.  Identification of intrinsic imaging phenotypes for breast cancer tumors: preliminary associations with gene expression profiles. , 2014, Radiology.

[13]  M. Giger,et al.  Computerized characterization of mammographic masses: analysis of spiculation. , 1994, Cancer letters.

[14]  O. Riesterer,et al.  Stability of radiomic features in CT perfusion maps , 2016, Physics in medicine and biology.

[15]  Lars J. Grimm,et al.  Computational approach to radiogenomics of breast cancer: Luminal A and luminal B molecular subtypes are associated with imaging features on routine breast MRI extracted using computer vision algorithms , 2015, Journal of magnetic resonance imaging : JMRI.

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

[17]  Lars J. Grimm,et al.  Relationships Between MRI Breast Imaging‐Reporting and Data System (BI‐RADS) Lexicon Descriptors and Breast Cancer Molecular Subtypes: Internal Enhancement is Associated with Luminal B Subtype , 2017, The breast journal.

[18]  Yuanjie Zheng,et al.  STEP: Spatiotemporal enhancement pattern for MR-based breast tumor diagnosis , 2009 .

[19]  C. Morris,et al.  Reliability of the manual ability classification system for children with cerebral palsy. , 2006, Developmental medicine and child neurology.

[20]  M. Giger,et al.  Automatic identification and classification of characteristic kinetic curves of breast lesions on DCE-MRI. , 2006, Medical physics.

[21]  W. Kaiser,et al.  Development, standardization, and testing of a lexicon for reporting contrast‐enhanced breast magnetic resonance imaging studies , 2001, Journal of magnetic resonance imaging : JMRI.

[22]  Hiroko Yamashita,et al.  Identifying Triple-Negative Breast Cancer Using Background Parenchymal Enhancement Heterogeneity on Dynamic Contrast-Enhanced MRI: A Pilot Radiomics Study , 2015, PloS one.

[23]  Maciej A Mazurowski,et al.  Algorithmic three-dimensional analysis of tumor shape in MRI improves prognosis of survival in glioblastoma: a multi-institutional study , 2017, Journal of Neuro-Oncology.

[24]  Catherine Klifa,et al.  Magnetic resonance imaging for secondary assessment of breast density in a high-risk cohort. , 2010, Magnetic resonance imaging.

[25]  Jennifer D. Brooks,et al.  Background parenchymal enhancement at breast MR imaging and breast cancer risk. , 2011, Radiology.

[26]  Maciej A Mazurowski,et al.  Interobserver variability in identification of breast tumors in MRI and its implications for prognostic biomarkers and radiogenomics. , 2016, Medical physics.

[27]  Maciej A Mazurowski,et al.  Radiogenomic analysis of breast cancer: luminal B molecular subtype is associated with enhancement dynamics at MR imaging. , 2014, Radiology.

[28]  Maria Bernathova,et al.  Diffusion‐weighted MRI of breast lesions: a prospective clinical investigation of the quantitative imaging biomarker characteristics of reproducibility, repeatability, and diagnostic accuracy , 2016, NMR in biomedicine.

[29]  Joseph O. Deasy,et al.  Breast cancer subtype intertumor heterogeneity: MRI‐based features predict results of a genomic assay , 2015, Journal of magnetic resonance imaging : JMRI.

[30]  K. Jung,et al.  Comparison of mammography, sonography, MRI and clinical examination in patients with locally advanced or inflammatory breast cancer who underwent neoadjuvant chemotherapy. , 2011, The British journal of radiology.

[31]  Maryellen L. Giger,et al.  A Fuzzy C-Means (FCM)-Based Approach for Computerized Segmentation of Breast Lesions in Dynamic Contrast-Enhanced MR Images1 , 2006 .

[32]  Bonnie N. Joe,et al.  Can signal enhancement ratio (SER) reduce the number of recommended biopsies without affecting cancer yield in occult MRI-detected lesions? , 2011, Academic radiology.

[33]  S. Walter,et al.  Sample size and optimal designs for reliability studies. , 1998, Statistics in medicine.

[34]  Ruijiang Li,et al.  Identifying relations between imaging phenotypes and molecular subtypes of breast cancer: Model discovery and external validation , 2017, Journal of magnetic resonance imaging : JMRI.

[35]  Kenneth G A Gilhuijs,et al.  Current clinical indications for magnetic resonance imaging of the breast , 2014, Journal of surgical oncology.

[36]  M. Giger,et al.  A fuzzy c-means (FCM)-based approach for computerized segmentation of breast lesions in dynamic contrast-enhanced MR images. , 2006, Academic radiology.

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

[38]  M. Sandelowski.  Sample size in qualitative research. , 1995, Research in nursing & health.

[39]  Jung Hyun Yoon,et al.  Breast parenchymal signal enhancement ratio at preoperative magnetic resonance imaging: association with early recurrence in triple-negative breast cancer patients , 2016, Acta radiologica.

[40]  Brian B. Avants,et al.  N4ITK: Improved N3 Bias Correction , 2010, IEEE Transactions on Medical Imaging.

[41]  A. Madabhushi,et al.  Computerized image analysis for identifying triple-negative breast cancers and differentiating them from other molecular subtypes of breast cancer on dynamic contrast-enhanced MR images: a feasibility study. , 2014, Radiology.

[42]  Maryellen L. Giger,et al.  Prediction of clinical phenotypes in invasive breast carcinomas from the integration of radiomics and genomics data , 2015, Journal of medical imaging.

[43]  Maciej A Mazurowski,et al.  Recurrence-free survival in breast cancer is associated with MRI tumor enhancement dynamics quantified using computer algorithms. , 2015, European journal of radiology.

[44]  M. Mazurowski.  Radiogenomics: what it is and why it is important. , 2015, Journal of the American College of Radiology : JACR.

[45]  Yan Chen,et al.  Accuracy of GE digital breast tomosynthesis versus supplementary mammographic views for diagnosis of screen-detected soft tissue breast lesions , 2015, Breast Cancer Research.

[46]  Breast imaging reporting and data system (BI-RADS) lexicon for breast MRI: interobserver variability in the description and assignment of BI-RADS category. , 2015, European journal of radiology.

[47]  S. Schneebaum,et al.  Tumor-to-breast volume ratio as measured on MRI: a possible predictor of breast-conserving surgery versus mastectomy. , 2014, The Israel Medical Association journal : IMAJ.

[48]  Lihua Li,et al.  A new quantitative image analysis method for improving breast cancer diagnosis using DCE-MRI examinations. , 2014, Medical physics.

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

[50]  P. Hartmann,et al.  Validation of nipple diameter and tongue movement measurements with B-mode ultrasound during breastfeeding. , 2010, Ultrasound in medicine & biology.

[51]  L. Bonomo,et al.  MRI accuracy in residual disease evaluation in breast cancer patients treated with neoadjuvant chemotherapy. , 2006, Clinical Radiology.

[52]  Wei Qian,et al.  Computer-aided breast MR image feature analysis for prediction of tumor response to chemotherapy. , 2015, Medical physics.

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

[54]  Shandong Wu,et al.  Quantitative assessment of background parenchymal enhancement in breast MRI predicts response to risk-reducing salpingo-oophorectomy: preliminary evaluation in a cohort of BRCA1/2 mutation carriers , 2015, Breast Cancer Research.

[55]  Lihua Li,et al.  Radiomic analysis of DCE-MRI for prediction of response to neoadjuvant chemotherapy in breast cancer patients. , 2017, European journal of radiology.

[56]  Michael D. Feldman,et al.  Pharmacokinetic Tumor Heterogeneity as a Prognostic Biomarker for Classifying Breast Cancer Recurrence Risk , 2015, IEEE Transactions on Biomedical Engineering.

[57]  Nikos Dimitropoulos,et al.  Multi-scaled morphological features for the characterization of mammographic masses using statistical classification schemes , 2007, Artif. Intell. Medicine.

[58]  Hon J. Yu,et al.  Quantitative analysis of lesion morphology and texture features for diagnostic prediction in breast MRI. , 2008, Academic radiology.

[59]  Andreas Makris,et al.  Inter‐ and intraobserver variability in the evaluation of dynamic breast cancer MRI , 2006, Journal of magnetic resonance imaging : JMRI.

[60]  M. Giger,et al.  Computerized interpretation of breast MRI: investigation of enhancement-variance dynamics. , 2004, Medical physics.

[61]  T. Helbich,et al.  Inter- and intra-observer agreement of BI-RADS-based subjective visual estimation of amount of fibroglandular breast tissue with magnetic resonance imaging: comparison to automated quantitative assessment , 2016, European Radiology.

[62]  L. Esserman,et al.  MRI measurements of breast tumor volume predict response to neoadjuvant chemotherapy and recurrence-free survival. , 2005, AJR. American journal of roentgenology.

[63]  Min-Ying Su,et al.  Comparison of breast density measured on MR images acquired using fat-suppressed versus nonfat-suppressed sequences. , 2011, Medical physics.