Fully Automated and Scalable Pipeline for Macaque Brain Registration

With the rapid development of various whole-brain ex vivo imaging technologies, there is a growing need to develop cross-modal 3D image registration methods to integrate multimodal imaging datasets. While comprehensive cross-modality registration tools such as D-LMBmap and mBrainAligner have been successfully implemented for mouse brains, macaque whole-brain registration presents unique challenges. These include more pronounced non-uniform deformations in larger ex vivo specimens, greater modality-specific contrast differences relative to standard space, and increased inter-individual variability. To address these challenges, we developed Macaca-Star, which incorporates deep learning models and self-individual MRI to tackle cross-modal and ex vivo sample deformation challenges in macaque whole-brain registration. Macaca-Star provides fully automated alignment of fMOST and 2D fluorescent slice images to the NMT MRI standard space, allowing for comprehensive integration of anterograde axonal projections and retrograde-traced neuronal soma profiles.