Initialize globally before acting locally: Enabling Landmark-free 3D US to MRI Registration
This addresses a practical bottleneck in medical image registration for clinicians, offering a more efficient alternative to landmark-based methods.
The paper tackles the problem of initializing 3D ultrasound to MRI registration without requiring tedious landmark annotations, by using Euclidean distance maps from coarse segmentations. It shows robustness on the RESECT dataset, providing initializations suitable for state-of-the-art deformable registration algorithms.
Registration of partial-view 3D US volumes with MRI data is influenced by initialization. The standard of practice is using extrinsic or intrinsic landmarks, which can be very tedious to obtain. To overcome the limitations of registration initialization, we present a novel approach that is based on Euclidean distance maps derived from easily obtainable coarse segmentations. We evaluate our approach quantitatively on the publicly available RESECT dataset and show that it is robust regarding overlap of target area and initial position. Furthermore, our method provides initializations that are suitable for state-of-the-art nonlinear, deformable image registration algorithm's capture ranges.