CLAIRE -- Parallelized Diffeomorphic Image Registration for Large-Scale Biomedical Imaging Applications

TL;DR

This paper introduces CLAIRE, a parallelized software for large-scale diffeomorphic image registration, demonstrating its ability to register extremely high-resolution biomedical images efficiently and analyzing the effects of downsampling on registration quality.

Contribution

The paper presents CLAIRE, a scalable multi-GPU registration algorithm capable of handling billion-voxel images and provides an extensive analysis of how downsampling affects registration accuracy.

Findings

Full-resolution registration can outperform downsampled registration in quality.

Downsampling from 1024^3 to 256^3 reduces Dice coefficient from 92% to 79%.

CLAIRE can register images of size 2816x3016x1162 in a few seconds.

Abstract

We study the performance of CLAIRE -- a diffeomorphic multi-node, multi-GPU image-registration algorithm, and software -- in large-scale biomedical imaging applications with billions of voxels. At such resolutions, most existing software packages for diffeomorphic image registration are prohibitively expensive. As a result, practitioners first significantly downsample the original images and then register them using existing tools. Our main contribution is an extensive analysis of the impact of downsampling on registration performance. We study this impact by comparing full-resolution registrations obtained with CLAIRE to lower-resolution registrations for synthetic and real-world imaging datasets. Our results suggest that registration at full resolution can yield a superior registration quality -- but not always. For example, downsampling a synthetic image from $1024^3$ to $256^3$ decreases the Dice coefficient from 92% to 79%. However, the differences are less pronounced for noisy or low-contrast high-resolution images. CLAIRE allows us not only to register images of clinically relevant size in a few seconds but also to register images at unprecedented resolution in a reasonable time. The highest resolution considered is CLARITY images of size $2816\times3016\times1162$. To the best of our knowledge, this is the first study on image registration quality at such resolutions.