Automated segmentation of mouse brain images using extended MRF

Abstract

We introduce an automated segmentation method, extended Markov random field (eMRF), to classify 21 neuroanatomical structures of mouse brain based on three dimensional (3D) magnetic resonance images (MRI). The image data are multispectral: T2-weighted, proton density-weighted, diffusion x, y and z weighted. Earlier research (Ali, A.A., Dale, A.M., Badea, A., Johnson, G.A., 2005. Automated segmentation of neuroanatomical structures in multispectral MR microscopy of the mouse brain. NeuroImage 27 (2), 425-435) successfully explored the use of MRF for mouse brain segmentation. In this research, we study the use of information generated from support vector machine (SVM) to represent the probabilistic information. Since SVM in general has a stronger discriminative power than the Gaussian likelihood method and is able to handle nonlinear classification problems, integrating SVM into MRF improved the classification accuracy. The eMRF employs the posterior probability distribution obtained from SVM to generate a classification based on the MR intensity. Secondly, the eMRF introduces a new potential function based on location information. Third, to maximize the classification performance, the eMRF uses the contribution weights optimally determined for each of the three potential functions: observation, location and contextual functions, which are traditionally equally weighted. We use the voxel overlap percentage and volume difference percentage to evaluate the accuracy of eMRF segmentation and compare the algorithm with three other segmentation methods--mixed ratio sampling SVM (MRS-SVM), atlas-based segmentation and MRF. Validation using classification accuracy indices between automatically segmented and manually traced data shows that eMRF outperforms other methods.

Figure 1

SVM binary classification problem (adopted from Vapnik, 1995). The solid line between the two dashed lines is the optimal hyperplane. The squares represent the samples from the positive class and the circles represent the samples from the negative class. The samples represented as the filled squares and the filled circles are the support vectors.

Figure 2

Comparison of the relative performances of the eMRF, MRS-SVM, atlas-based segmentation and MRF methods based on the voxel overlap percent - VOP (top) and the volume difference percent -VDP (bottom) indices

Figure 3

Coronal slices through the labeled brain at the level of anterior hippocampus and third ventricle (upper row), and pons and substantia nigra (lower row) show in a qualitative manner the relative superiority of eMRF compared to MRS-SVM. Note that eMRF segmentation better preserved the shapes of striatum and corpus callosum (as seen in the manual labels), compared to MRS-SVM; and also that eMRF was able to segment a small CSF filled region in the center of PAG, while MRS-SVM missed it.