Upscaling Mouse Spinal Cord Atlas for 3D Alignment

Atlas plays a crucial role in the field of neuroscience. While brain atlases have been established in 2D and 3D models in recent years, the exploration and research on spinal cord atlases are significantly lacking. The currently widely used 2D mouse spinal cord atlas provides detailed information on 2D sections and has become a mature alignment template for data analysis. However, this cannot replace the precision and visual experience offered by a 3D atlas. Nevertheless, a 3D mouse spinal cord atlas remains unavailable to the community. Obtaining a 3D mouse spinal cord atlas through traditional experimental methods requires high labor and time costs. Therefore, in this study, we reconstructed a 3D mouse spinal cord atlas using modeling software and algorithms based on region-specific interpolation of the annotated 2D mouse spinal cord atlas, resulting in a 3D atlas with high credibility and resolution. The region-specific interpolation was achieved using a bridging loop algorithm. After voxelizing the resultant model, we obtained a high-resolution, smoothed 3D mouse spinal cord atlas. By extracting voxel points through code and optimizing the smooth structures, we could generate a 2D atlas with any arbitrary resolution along the Z-axis. We now provide the 3D atlas as well as a 2D mouse spinal cord atlas enhanced by five-fold resolution. This new 3D spinal cord atlas facilitates standardized image registration for cross-comparison purposes. The method proposed in this paper could also serve as a reference for enhancing the resolution of other 2D atlases in specific directions.