Projects:RegistrationRegularization

We propose a unified framework for computing atlases from manually labeled data sets at various degrees of “sharpness” and the joint registration and segmentation of a new brain with these atlases. In non-rigid registration, the tradeoff between warp regularization and image fidelity is typically set empirically. In segmentation, this leads to a probabilistic atlas of arbitrary “sharpness”: weak regularization results in well-aligned training images, producing a “sharp” atlas; strong regularization yields a “blurry” atlas. We study the effects of this tradeoff in the context of cortical surface parcellation, but the framework applies to volume registration as well.

Results

We compare three special cases of our framework, namely: progressive registration-segmentation of a new brain to increasingly “sharp” atlases using increasingly flexible warps; secondly, progressive registration to a single atlas with increasingly flexible warps; and thirdly, registration to a single atlas with fixed constrained warps. We find that the optimal parcellation algorithm in all three cases yield a statistically significant improvement over a state-of-the-art benchmark parcellation algorithm. The optimal algorithms correspond to a unique balance between atlas “sharpness” and warp regularization.