A Multichannel Markov Random Field Framework for Tumor Segmentation With an Application to Classification of Gene Expression-Based Breast Cancer Recurrence Risk

• Published in: IEEE transactions on medical imaging Vol. 32; no. 4; pp. 637 - 648
• Main Authors: Ashraf, A. B., Gavenonis, S. C., Daye, D., Mies, C., Rosen, M. A., Kontos, D.
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.04.2013; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Area Under Curve; Belief propagation; Breast cancer; Breast cancer recurrence prediction; breast dynamic contrast enhanced (DCE) magnetic resonance imaging (MRI); Breast Neoplasms - genetics; Breast Neoplasms - metabolism; Breast Neoplasms - pathology; breast tumor segmentation; Computational Biology; Equations; Female; Gene expression; Gene Expression Profiling; Humans; Image Interpretation, Computer-Assisted - methods; Image segmentation; Inference algorithms; Kinetic theory; Kinetics; Magnetic Resonance Imaging - methods; Markov Chains; Mathematical model; Neoplasm Recurrence, Local - genetics; Neoplasm Recurrence, Local - metabolism; Neoplasm Recurrence, Local - pathology; Predictive Value of Tests; Reproducibility of Results; ROC Curve; Studies; Support Vector Machine; tumor characterization; Tumors
• ISSN: 0278-0062; 1558-254X
• DOI: 10.1109/TMI.2012.2219589

We present a methodological framework for multichannel Markov random fields (MRFs). We show that conditional independence allows loopy belief propagation to solve a multichannel MRF as a single channel MRF. We use conditional mutual information to search for features that satisfy conditional independence assumptions. Using this framework we incorporate kinetic feature maps derived from breast dynamic contrast enhanced magnetic resonance imaging as observation channels in MRF for tumor segmentation. Our algorithm based on multichannel MRF achieves an receiver operating characteristic area under curve (AUC) of 0.97 for tumor segmentation when using a radiologist's manual delineation as ground truth. Single channel MRF based on the best feature chosen from the same pool of features as used by the multichannel MRF achieved a lower AUC of 0.89. We also present a comparison against the well established normalized cuts segmentation algorithm along with commonly used approaches for breast tumor segmentation including fuzzy C-means (FCM) and the more recent method of running FCM on enhancement variance features (FCM-VES). These previous methods give a lower AUC of 0.92, 0.88, and 0.60, respectively. Finally, we also investigate the role of superior segmentation in feature extraction and tumor characterization. Specifically, we examine the effect of improved segmentation on predicting the probability of breast cancer recurrence as determined by a validated tumor gene expression assay. We demonstrate that an support vector machine classifier trained on kinetic statistics extracted from tumors as segmented by our algorithm gives a significant improvement in distinguishing between women with high and low recurrence risk, giving an AUC of 0.88 as compared to 0.79, 0.76, 0.75, and 0.66 when using normalized cuts, single channel MRF, FCM, and FCM-VES, respectively, for segmentation.