Matching of diffusion tensor images using Gabor features

This paper presents a novel method for feature-based matching of diffusion tensor images using the complete tensor information available at each voxel rather than limited scalar parameters such as the fractional anisotropy. In our method, we characterize each voxel by a rich rotationally invariant feature vector defined using Gabor filters. In order to obtain these features, the gabor filters are evaluated at multiple scales and frequencies and are oriented along the dominant direction of the tensors in a neighborhood around the voxel under consideration. The feature is able to obtain a fine to coarse description of each voxel and fully accounts for the highly oriented nature of the tensor data. The proposed matching paradigm based on these Gabor features has been tested on simulated and real images and produces good correspondences.

[1]  John Daugman,et al.  Neural networks for image transformation, analysis, and compression , 1988, Neural Networks.

[2]  Dinggang Shen,et al.  Segmentation of prostate boundaries from ultrasound images using statistical shape model , 2003, IEEE Transactions on Medical Imaging.

[3]  B. S. Manjunath,et al.  Texture Features for Browsing and Retrieval of Image Data , 1996, IEEE Trans. Pattern Anal. Mach. Intell..

[4]  Dinggang Shen,et al.  Generalized Affine Invariant Image Normalization , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[5]  William E. Higgins,et al.  Texture Segmentation using 2-D Gabor Elementary Functions , 1994, IEEE Trans. Pattern Anal. Mach. Intell..

[6]  James C. Gee,et al.  Spatial transformations of diffusion tensor magnetic resonance images , 2001, IEEE Transactions on Medical Imaging.

[7]  P. Basser,et al.  MR diffusion tensor spectroscopy and imaging. , 1994, Biophysical journal.

[8]  P. Basser,et al.  Diffusion tensor MR imaging of the human brain. , 1996, Radiology.

[9]  Jundong Liu,et al.  Local frequency representations for robust multimodal image registration , 2002, IEEE Transactions on Medical Imaging.