Fast interactive multi-region cardiac segmentation with linearly ordered labels

We present a novel and fast interactive approach to multi-modality cardiac image segmentation, which employs the linearly ordered surfaces as an additional constraint. We show using such a geometrical constraint helps to significantly reduce user interaction and improve the accuracy of segmentation results at the same time. We solve the proposed multiregion segmentation problem with the order constraints by means of convex optimization, resulting in a fast and reliable flow maximization approach which implicitly embeds the linear order prior without introducing extra computation load. In this regard, a new fully parallelized continuous max-flow algorithm is proposed and implemented using GPGPU to segment a 3D volume within one second. We demonstrate our results over pathological trans-esophageal echocardiogram, cardiac CT and delayed enhancement MRI data sets.

[1]  Terry M. Peters,et al.  A Convex Max-Flow Segmentation of LV Using Subject-Specific Distributions on Cardiac MRI , 2011, IPMI.

[2]  Olga Veksler,et al.  Star Shape Prior for Graph-Cut Image Segmentation , 2008, ECCV.

[3]  Milan Sonka,et al.  4-D Cardiac MR Image Analysis: Left and Right Ventricular Morphology and Function , 2010, IEEE Transactions on Medical Imaging.

[4]  Guido Gerig,et al.  User-guided 3D active contour segmentation of anatomical structures: Significantly improved efficiency and reliability , 2006, NeuroImage.

[5]  Xue-Cheng Tai,et al.  A study on continuous max-flow and min-cut approaches , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[6]  Shuo Li,et al.  Left Ventricle Segmentation via Graph Cut Distribution Matching , 2009, MICCAI.

[7]  Olga Veksler,et al.  Tiered scene labeling with dynamic programming , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[8]  Hiroshi Ishikawa,et al.  Exact Optimization for Markov Random Fields with Convex Priors , 2003, IEEE Trans. Pattern Anal. Mach. Intell..

[9]  Ghassan Hamarneh,et al.  Active Learning for Interactive 3D Image Segmentation , 2011, MICCAI.

[10]  Daniel Cremers,et al.  A convex approach for computing minimal partitions , 2008 .

[11]  Marie-Pierre Jolly,et al.  Interactive graph cuts for optimal boundary & region segmentation of objects in N-D images , 2001, Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001.

[12]  Yuri Boykov,et al.  Globally optimal segmentation of multi-region objects , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[13]  Terry M. Peters,et al.  Real-time segmentation in 4D ultrasound with continuous max-flow , 2012, Medical Imaging.

[14]  Mila Nikolova,et al.  Algorithms for Finding Global Minimizers of Image Segmentation and Denoising Models , 2006, SIAM J. Appl. Math..

[15]  N. Paragios.  A level set approach for shape-driven segmentation and tracking of the left ventricle , 2003, IEEE Transactions on Medical Imaging.

[16]  Wolfgang Birkfellner,et al.  Applied Medical Image Processing: A Basic Course , 2010 .

[17]  Xiaodong Wu,et al.  Simultaneous searching of globally optimal interacting surfaces with shape priors , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.