Multi-modal Multi-Atlas Segmentation using Discrete Optimisation and Self-Similarities

This work presents the application of a discrete medical image registration framework to multi-organ segmentation in different modalities. The algorithm works completely automatically and does not have to be tuned specifically for different datasets. A robust similarity measure, using the local self-similarity context (SSC), is employed and shown to outperform other commonly used metrics. Both affine and deformable registration are driven by a dense displacement sampling (deeds) strategy. The smoothness of displacements is enforced by inference on a Markov random field (MRF), using a tree approximation for computational efficiency. Consensus segmentations for unseen test images of the VISCERAL Anatomy 3 data are found by majority voting.