Uncertainty driven probabilistic voxel selection for image registration

TL;DR

This paper introduces a Bayesian probabilistic voxel selection method for medical image registration that adaptively samples voxels based on uncertainty, achieving high accuracy with minimal voxel usage in time-sensitive scenarios.

Contribution

It proposes a novel probabilistic voxel sampling strategy using a Bayesian framework that balances exploration and exploitation during registration.

Findings

Effective voxel sampling with less than 1% of total voxels.

Maintains registration accuracy and low failure rate.

Balances robustness and accuracy through probabilistic sampling.

Abstract

This paper presents a novel probabilistic voxel selection strategy for medical image registration in time-sensitive contexts, where the goal is aggressive voxel sampling (e.g. using less than 1% of the total number) while maintaining registration accuracy and low failure rate. We develop a Bayesian framework whereby, first, a voxel sampling probability field (VSPF) is built based on the uncertainty on the transformation parameters. We then describe a practical, multi-scale registration algorithm, where, at each optimization iteration, different voxel subsets are sampled based on the VSPF. The approach maximizes accuracy without committing to a particular fixed subset of voxels. The probabilistic sampling scheme developed is shown to manage the tradeoff between the robustness of traditional random voxel selection (by permitting more exploration) and the accuracy of fixed voxel selection (by permitting a greater proportion of informative voxels).