Cross-Sim-NGF: FFT-Based Global Rigid Multimodal Alignment of Image Volumes using Normalized Gradient Fields
This addresses the need for fast and accurate automated alignment of medical image volumes across different modalities, such as brain scans, which is incremental as it improves upon existing global optimization techniques.
The paper tackled the problem of global rigid multimodal 3D image alignment by proposing an FFT-based method using normalized gradient fields, which outperformed four reference methods by a large margin and reduced computation time by over three orders of magnitude, achieving alignment in about 40 seconds for 3.4Mvoxel volumes.
Multimodal image alignment involves finding spatial correspondences between volumes varying in appearance and structure. Automated alignment methods are often based on local optimization that can be highly sensitive to their initialization. We propose a global optimization method for rigid multimodal 3D image alignment, based on a novel efficient algorithm for computing similarity of normalized gradient fields (NGF) in the frequency domain. We validate the method experimentally on a dataset comprised of 20 brain volumes acquired in four modalities (T1w, Flair, CT, [18F] FDG PET), synthetically displaced with known transformations. The proposed method exhibits excellent performance on all six possible modality combinations, and outperforms all four reference methods by a large margin. The method is fast; a 3.4Mvoxel global rigid alignment requires approximately 40 seconds of computation, and the proposed algorithm outperforms a direct algorithm for the same task by more than three orders of magnitude. Open-source implementation is provided.