Automatic breast DCE-MRI segmentation using compound morphological operations

Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is an evolving tool for determining breast diseases. Assisted with computer-aided methods (CAD), DCE-MRI demonstrates its efficiency on breast cancer diagnosis. A crucial step for automatic DCE-MRI analysis is to separate breast regions from the noise-corrupted background on DCE-MRI images. In this paper, we presented an automatic algorithm for breast segmentation of sagittal DCE-MRI sequences. To separate the background, the images were converted into binary images. Multiple morphological operations were then applied to remove the undesired noise and artifacts. To separate the breast regions from the chests, a chest contour mask was generated based on the overlap of the whole sequences. The mask was applied on each frame to exclude the chest region. The method was validated against manual segmentation and evaluated with different performance indexes. Our results demonstrate a high accuracy of the method compared with the manual segmentations.

[1]  Onur Osman,et al.  Breast MR segmentation and lesion detection with cellular neural networks and 3D template matching , 2008, Comput. Biol. Medicine.

[2]  Tim W. Nattkemper,et al.  An Adaptive Tissue Characterisation Network for Model-Free Visualisation of Dynamic Contrast-Enhanced Magnetic Resonance Image Data , 2005 .

[3]  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 .

[4]  Lia Morra,et al.  A fully automatic lesion detection method for DCE-MRI fat-suppressed breast images , 2009, Medical Imaging.

[5]  Stuart Crozier,et al.  Edge intensity normalization as a bias field correction during balloon snake segmentation of breast MRI , 2008, 2008 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[6]  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.

[7]  L. Turnbull.  Dynamic contrast‐enhanced MRI in the diagnosis and management of breast cancer , 2009, NMR in biomedicine.

[8]  Chao Lu,et al.  DCE-MRI Segmentation and Motion Correction Based on Active Contour Model and Forward Mapping , 2006, Seventh ACIS International Conference on Software Engineering, Artificial Intelligence, Networking, and Parallel/Distributed Computing (SNPD'06).

[9]  Joseph M. Reinhardt,et al.  Classification of breast MRI lesions using a backpropagation neural network (BNN) , 2004, 2004 2nd IEEE International Symposium on Biomedical Imaging: Nano to Macro (IEEE Cat No. 04EX821).

[10]  A. Bert,et al.  A fully automatic algorithm for segmentation of the breasts in DCE-MR images , 2010, 2010 Annual International Conference of the IEEE Engineering in Medicine and Biology.

[11]  H. O. Gulcur,et al.  An Adaptive Approach to the Segmentation of DCE-MR Images of the Breast: Comparison with Classical Thresholding Algorithms , 2007, 2007 IEEE Symposium on Computational Intelligence in Image and Signal Processing.