PDE-LDDMM meets NODEs: Introducing neural ordinary differential equation solvers in PDE-constrained Large Deformation Diffeomorphic Metric Mapping
Year: 2024
Type: article
Abstract: Non-rigid image registration is a crucial task in various medical applications, allowing thealignment of images with complex spatial or temporal variations. This paper introduces NODEO-LDDMMand NODEO-PDE-LDDMM, two innovative deep-learning-based approaches that bridge the gap between LargeDeformation Diffeomorphic Metric Mapping (LDDMM) and neural ordinary differential equations (NODEs).LDDMM and PDE-LDDMM offer mathematically well-established formulations for diffeomorphic registration,while NODEs provide the flexibility of deep-learning in the solution of the ODEs involved in bothmethods.Both NODEO-LDDMM and NODEO-PDE-LDDMM include the strengths of deep-learning into LDDMM, enabling arobust optimization with a good balance between accuracy and transformation smoothness in theirsolutions.Our proposed methods reached or outperformed their traditional counterparts and the nearlydiffeomorphic deep-learning-based approaches selected as benchmarks.This work contributes to advancing non-rigid image registration techniques, with a methodologysuited to overcome some of the limitations of deep-learning in medical image registration.
Source: Journal of Computational Science
Author: Monica Hernandez
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $2,890
OpenAlex ID: W4405430224