MUSE: MUlti-atlas region Segmentation utilizing Ensembles of registration algorithms and parameters, and locally optimal atlas selection

Jimit Doshi, Güray Erus, Yangming Ou, Susan M. Resnick, Ruben C. Gur, Raquel E. Gur, Theodore D. Satterthwaite, Susan Furth, Christos Davatzikos

Abstract: Highlights•A new multiatlas segmentation framework using a broad ensemble of labeled templates•Combines different atlases, warping algorithms, and regularization parameters•Uses an adaptive fusion strategy through local similarity weighting and intensity based refinement•Ensemble approach provides robustness to image variations and produces accurate segmentations.