Fast and Globally Convex Multiphase Active Contours for Brain MRI Segmentation

Abstract:

Multiphase active contour based models are useful in identifying multiple regions with spatial consistency but varying characteristics such as the mean intensities of regions. Segmenting brain magnetic resonance images (MRIs) using a multiphase approach is useful to differentiate white and gray matter tissue for anatomical, functional and disease studies. Multiphase active contour methods are superior to other approaches due to their topological flexibility, accurate boundaries, robustness to image variations and adaptive energy functionals. Globally convex methods are furthermore initialization independent. We extend the relaxed globally convex Chan and Vese two-phase piecewise constant energy minimization formulation to the multiphase domain and prove the existence of a global minimizer. An efficient dual minimization implementation of our binary partitioning function model accurately describes disjoint regions using stable segmentations by avoiding local minima solutions. Experimental results indicate that the proposed approach provides consistently better accuracy than other related multiphase active contour algorithms using four different error metrics (Dice, Rand Index, Global Consistency Error and Variation of Information) even under severe noise, intensity inhomogeneities, and partial volume effects in MRI imagery.

Example Segmentation Results

Our fast and automatic four phase image segmentation scheme provides a better segmentations for brain MRI images, it differentiates the gray matter from the surrounding white region clearly. (a) Input image, coronal slice from a normal brain MR imagery. (b) & (c) Show final binary segmentations obtained by thresholding the relaxed functions u1, u2 at 0.5, (d) Final segmentation result showing the contours superimposed on the input image. (e) Color coded visualization of the obtained segmentation result.

Acknowledgment:

^*This work was done while the author was at the IPAM, University of California Los Angeles, CA, USA. The author thanks the IPAM institute for their great hospitality and support during the visit.

Back to MRI Projects Main Page. Back to Research Page.