Learning accurate rigid registration for longitudinal brain MRI from synthetic data
This work addresses the critical need for accurate longitudinal brain MRI registration, which is incremental as it builds on an existing framework for anatomy-aware affine registration.
The paper tackled the problem of achieving precise alignment in longitudinal (within-subject) brain MRI registration by proposing a model optimized for rigid registration, resulting in more accurate rigid transforms than previous cross-subject networks and robust performance across MRI contrasts.
Rigid registration aims to determine the translations and rotations necessary to align features in a pair of images. While recent machine learning methods have become state-of-the-art for linear and deformable registration across subjects, they have demonstrated limitations when applied to longitudinal (within-subject) registration, where achieving precise alignment is critical. Building on an existing framework for anatomy-aware, acquisition-agnostic affine registration, we propose a model optimized for longitudinal, rigid brain registration. By training the model with synthetic within-subject pairs augmented with rigid and subtle nonlinear transforms, the model estimates more accurate rigid transforms than previous cross-subject networks and performs robustly on longitudinal registration pairs within and across magnetic resonance imaging (MRI) contrasts.