Multilevel analysis of spatiotemporal association features for differentiation of tumor enhancement patterns in breast DCE-MRI.

PURPOSE Analyzing spatiotemporal enhancement patterns is an important task for the differential diagnosis of breast tumors in dynamic contrast-enhanced MRI (DCE-MRI), and yet remains challenging because of complexities in analyzing the time-series of three-dimensional image data. The authors propose a novel approach to breast MRI computer-aided diagnosis (CAD) using a multilevel analysis of spatiotemporal association features for tumor enhancement patterns in DCE-MRI. METHODS A database of 171 cases consisting of 111 malignant and 60 benign tumors was used. Time-series contrast-enhanced MR images were obtained from two different types of MR scanners and protocols. The images were first registered for motion compensation, and then tumor regions were segmented using a fuzzy c-means clustering-based method. Spatiotemporal associations of tumor enhancement patterns were analyzed at three levels: Mapping of pixelwise kinetic features within a tumor, extraction of spatial association features from kinetic feature maps, and extraction of kinetic association features at the spatial feature level. A total of 84 initial features were extracted. Predictable values of these features were evaluated with an area under the ROC curve, and were compared between the spatiotemporal association features and a subset of simple form features which do not reflect spatiotemporal association. Several optimized feature sets were identified among the spatiotemporal association feature group or among the simple feature group based on a feature ranking criterion using a support vector machine based recursive feature elimination algorithm. A least-squares support vector machine (LS-SVM) classifier was used for tumor differentiation and the performances were evaluated using a leave-one-out testing. RESULTS Predictable values of the extracted single features ranged in 0.52-0.75. By applying multilevel analysis strategy, the spatiotemporal association features became more informative in predicting tumor malignancy, which was shown by a statistical testing in ten spatiotemporal association features. By using a LS-SVM classifier with the optimized second and third level feature set, the CAD scheme showed Az of 0.88 in classification of malignant and benign tumors. When this performance was compared to the same LS-SVM classifier with simple form features which do not reflect spatiotemporal association, there was a statistically significant difference (0.88 vs 0.79, p <0.05), suggesting that the multilevel analysis strategy yields a significant performance improvement. CONCLUSIONS The results suggest that the multilevel analysis strategy characterizes the complex tumor enhancement patterns effectively with the spatiotemporal association features, which in turn leads to an improved tumor differentiation. The proposed CAD scheme has a potential for improving diagnostic performance in breast DCE-MRI.

[1]  G. Marchal,et al.  Multi-modal volume registration by maximization of mutual information , 1997 .

[2]  Alain Rakotomamonjy,et al.  Variable Selection Using SVM-based Criteria , 2003, J. Mach. Learn. Res..

[3]  R. Fimmers,et al.  Mammography, breast ultrasound, and magnetic resonance imaging for surveillance of women at high familial risk for breast cancer. , 2005, Journal of clinical oncology : official journal of the American Society of Clinical Oncology.

[4]  Dinggang Shen,et al.  STEP: SPATIAL-TEMPORAL ENHANCEMENT PATTERN, FOR MR-BASED BREAST TUMOR DIAGNOSIS , 2007, 2007 4th IEEE International Symposium on Biomedical Imaging: From Nano to Macro.

[5]  Jason Weston,et al.  Gene Selection for Cancer Classification using Support Vector Machines , 2002, Machine Learning.

[6]  Tanja Alderliesten,et al.  Validation of Semiautomatic Measurement of the Extent of Breast Tumors Using Contrast-Enhanced Magnetic Resonance Imaging , 2007, Investigative radiology.

[7]  Brian D. Ripley,et al.  Pattern Recognition and Neural Networks , 1996 .

[8]  Johan A. K. Suykens,et al.  Least Squares Support Vector Machine Classifiers , 1999, Neural Processing Letters.

[9]  J. Hanley,et al.  A method of comparing the areas under receiver operating characteristic curves derived from the same cases. , 1983, Radiology.

[10]  F. Kelcz,et al.  Critical role of spatial resolution in dynamic contrast‐enhanced breast MRI , 2001, Journal of magnetic resonance imaging : JMRI.

[11]  Frederick Kelcz,et al.  Clinical testing of high-spatial-resolution parametric contrast-enhanced MR imaging of the breast. , 2002, AJR. American journal of roentgenology.

[12]  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 .

[13]  G Brix,et al.  Classification of signal-time curves from dynamic MR mammography by neural networks. , 2001, Magnetic resonance imaging.

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

[15]  E A Sickles,et al.  Dynamic high-spatial-resolution MR imaging of suspicious breast lesions: diagnostic criteria and interobserver variability. , 2000, AJR. American journal of roentgenology.

[16]  Laura Cortesi,et al.  Multicenter comparative multimodality surveillance of women at genetic-familial high risk for breast cancer (HIBCRIT study): interim results. , 2007, Radiology.

[17]  Dov,et al.  Response of MCF 7 Human Breast Cancer to Tamoxifen : Evaluation by the Three-Time-Point , Contrast-enhanced Magnetic Resonance Imaging Method 1 , 2005 .

[18]  Martin J. McKeown,et al.  Spatial Characterization of fMRI Activation Maps Using Invariant 3-D Moment Descriptors , 2009, IEEE Transactions on Medical Imaging.

[19]  Joan Leach,et al.  Expert heads a'rolling Melissa Leach Ian , 2005, The Lancet.

[20]  N. Hylton,et al.  Diagnostic architectural and dynamic features at breast MR imaging: multicenter study. , 2006, Radiology.

[21]  S G Orel,et al.  Differentiating benign from malignant enhancing lesions identified at MR imaging of the breast: are time-signal intensity curves an accurate predictor? , 1999, Radiology.

[22]  L. Turnbull,et al.  Dynamic contrast‐enhanced MRI in the differentiation of breast tumors: User‐defined versus semi‐automated region‐of‐interest analysis , 1999, Journal of magnetic resonance imaging : JMRI.

[23]  Xuan Guo,et al.  Three-Dimensional Moment Invariants under Rigid Transformation , 1993, CAIP.

[24]  Ying Lu,et al.  Kinetic assessment of breast tumors using high spatial resolution signal enhancement ratio (SER) imaging , 2007, Magnetic resonance in medicine.

[25]  B. Szabó,et al.  Dynamic MR imaging of the breast: Analysis of kinetic and morphologic diagnostic criteria , 2003, Acta radiologica.

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

[27]  P. Chappuis,et al.  Re: Magnetic resonance imaging and mammography in women with a hereditary risk of breast cancer. , 2001, Journal of the National Cancer Institute.

[28]  Wendy B DeMartini,et al.  Breast MR imaging: computer-aided evaluation program for discriminating benign from malignant lesions. , 2007, Radiology.

[29]  C. Kuhl,et al.  Dynamic breast MR imaging: are signal intensity time course data useful for differential diagnosis of enhancing lesions? , 1999, Radiology.

[30]  Mieczyslaw Gajda,et al.  Application of computer-aided diagnosis (CAD) in MR-mammography (MRM): do we really need whole lesion time curve distribution analysis? , 2009, Academic radiology.

[31]  Scott Fields,et al.  Mapping pathophysiological features of breast tumors by MRI at high spatial resolution , 1997, Nature Medicine.

[32]  A Horsman,et al.  Dynamic MR imaging of invasive breast cancer: correlation with tumour grade and other histological factors. , 1997, The British journal of radiology.

[33]  R. Warren,et al.  Screening with magnetic resonance imaging and mammography of a UK population at high familial risk of breast cancer: a prospective multicentre cohort study (MARIBS) , 2005, The Lancet.

[34]  Torsten Rohlfing,et al.  Volume-preserving nonrigid registration of MR breast images using free-form deformation with an incompressibility constraint , 2003, IEEE Transactions on Medical Imaging.

[35]  David J. Hawkes,et al.  Does registration improve the performance of a computer aided diagnosis system for dynamic contrast-enhanced MR mammography? , 2006, 3rd IEEE International Symposium on Biomedical Imaging: Nano to Macro, 2006..

[36]  Lina Arbash Meinel,et al.  Breast MRI lesion classification: Improved performance of human readers with a backpropagation neural network computer‐aided diagnosis (CAD) system , 2007, Journal of magnetic resonance imaging : JMRI.

[37]  M. Leach.  Breast cancer screening in women at high risk using MRI , 2009, NMR in biomedicine.

[38]  L. Liberman,et al.  Breast cancer screening with MRI--what are the data for patients at high risk? , 2004, The New England journal of medicine.

[39]  M D Schnall,et al.  Update of breast MR imaging architectural interpretation model. , 2001, Radiology.

[40]  Daniel Rueckert,et al.  Nonrigid registration using free-form deformations: application to breast MR images , 1999, IEEE Transactions on Medical Imaging.

[41]  A Horsman,et al.  Dynamic MRI of invasive breast cancer: assessment of three region-of-interest analysis methods. , 1997, Journal of computer assisted tomography.

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

[43]  Lubomir M. Hadjiiski,et al.  Treatment response assessment of breast masses on dynamic contrast-enhanced magnetic resonance scans using fuzzy c-means clustering and level set segmentation. , 2009, Medical physics.

[44]  Anne L. Martel,et al.  Classification of Dynamic Contrast-Enhanced Magnetic Resonance Breast Lesions by Support Vector Machines , 2008, IEEE Transactions on Medical Imaging.

[45]  N M Hylton,et al.  Vascularity assessment of breast lesions with gadolinium-enhanced MR imaging. , 1999, Magnetic resonance imaging clinics of North America.

[46]  Vladimir Vapnik,et al.  Statistical learning theory , 1998 .

[47]  T Yasunaga,et al.  Evaluation of tumor angiogenesis using dynamic enhanced magnetic resonance imaging: comparison of plasma vascular endothelial growth factor, hemodynamic, and pharmacokinetic parameters , 2004, Acta radiologica.

[48]  U. Bick,et al.  Differentiation between benign and malignant findings on MR-mammography: usefulness of morphological criteria , 2001, European Radiology.