FastReg: Fast Non-Rigid Registration via Accelerated Optimisation on the Manifold of Diffeomorphisms
This addresses the need for faster and more efficient image registration in medical imaging, particularly for applications like brain and abdominal scans, though it appears incremental as it builds on existing optical flow and gradient flow techniques.
The paper tackles the problem of slow diffeomorphic non-rigid registration for medical images by introducing an accelerated optimization method on the manifold of diffeomorphisms, achieving registration speeds orders of magnitude faster than previous approaches for brain MRI and abdominal CT scans.
We present an implementation of a new approach to diffeomorphic non-rigid registration of medical images. The method is based on optical flow and warps images via gradient flow with the standard $L^2$ inner product. To compute the transformation, we rely on accelerated optimisation on the manifold of diffeomorphisms. We achieve regularity properties of Sobolev gradient flows, which are expensive to compute, owing to a novel method of averaging the gradients in time rather than space. We successfully register brain MRI and challenging abdominal CT scans at speeds orders of magnitude faster than previous approaches. We make our code available in a public repository: https://github.com/dgrzech/fastreg