RealKeyMorph: Keypoints in Real-world Coordinates for Resolution-agnostic Image Registration

TL;DR

RealKeyMorph (RKM) is a novel resolution-agnostic image registration method that directly predicts keypoints in real-world coordinates, avoiding resampling artifacts and improving registration across varying image resolutions.

Contribution

RKM extends KeyMorph by operating in real-world coordinates, enabling resolution-agnostic registration without resampling, applicable to 2D and 3D medical images.

Findings

RKM outperforms resampling-based methods in registration accuracy.

Effective on both 2D abdominal MRI stacks and 3D brain datasets.

Demonstrates robustness across different resolutions.

Abstract

Many real-world settings require registration of a pair of medical images that differ in spatial resolution, which may arise from differences in image acquisition parameters like pixel spacing, slice thickness, and field-of-view. However, all previous machine learning-based registration techniques resample images onto a fixed resolution. This is suboptimal because resampling can introduce artifacts due to interpolation. To address this, we present RealKeyMorph (RKM), a resolution-agnostic method for image registration. RKM is an extension of KeyMorph, a registration framework which works by training a network to learn corresponding keypoints for a given pair of images, after which a closed-form keypoint matching step is used to derive the transformation that aligns them. To avoid resampling and enable operating on the raw data, RKM outputs keypoints in real-world coordinates of the scanner. To do this, we leverage the affine matrix produced by the scanner (e.g., MRI machine) that encodes the mapping from voxel coordinates to real world coordinates. By transforming keypoints into real-world space and integrating this into the training process, RKM effectively enables the extracted keypoints to be resolution-agnostic. In our experiments, we demonstrate the advantages of RKM on the registration task for orthogonal 2D stacks of abdominal MRIs, as well as 3D volumes with varying resolutions in brain datasets.