Enhancing Label-Driven Deep Deformable Image Registration with Local Distance Metrics for State-of-the-Art Cardiac Motion Tracking
This work addresses the challenge of achieving state-of-the-art accuracy in medical image registration for cardiac applications, representing an incremental improvement over existing deep learning approaches.
The paper tackled the problem of improving accuracy in deformable image registration for cardiac motion tracking by combining global semantic information from segmentation labels with local distance metrics, achieving a Dice score of 86.5%, which is higher than classic methods (79.0%) and label-driven deep learning frameworks (83.4%).
While deep learning has achieved significant advances in accuracy for medical image segmentation, its benefits for deformable image registration have so far remained limited to reduced computation times. Previous work has either focused on replacing the iterative optimization of distance and smoothness terms with CNN-layers or using supervised approaches driven by labels. Our method is the first to combine the complementary strengths of global semantic information (represented by segmentation labels) and local distance metrics that help align surrounding structures. We demonstrate significant higher Dice scores (of 86.5\%) for deformable cardiac image registration compared to classic registration (79.0\%) as well as label-driven deep learning frameworks (83.4\%).