Mapping Histological Slice Sequences to the Allen Mouse Brain Atlas without 3D Reconstruction
News
4/24/2022
added a preprocessing script to downsize a tif image file and manually remove background/easily-replaced region in a slice.
added a preprocessing script to select sectional atlas.
4/23/2022 added a script demo to remove the background in a slice for preprocessing.
12/06/2021 released a step by step version of the code with which neuroscientists can verify and select the best cutting angle, check matched and registered slices. Screenshots of stepwise instructions are included in the "screenshot instructions" folder. DM if you have any questions.
09/05/2021 added the rotate_annotation.m file for 2D image registration.
09/07/2021 added additional missing files to make sure the 2D registration demo would be running successfully
Abstract
Histological brain slices are widely used in neuroscience to study the anatomical organization of neural circuits. Systematic and accurate comparisons of anatomical data from multiple brains, especially from different studies, can benefit tremendously from registering histological slices onto a common reference atlas. Most existing methods rely on an initial reconstruction of the volume before registering it to a reference atlas. Because these slices are prone to distortions during the sectioning process and often sectioned with non-standard angles, reconstruction is challenging and often inaccurate. Here we describe a framework that maps each slice to its corresponding plane in the Allen Mouse Brain Atlas (2015) to build a plane-wise mapping and then perform 2D nonrigid registration to build a pixel-wise mapping. We use the L2 norm of the histogram of oriented gradients of two patches as the similarity metric for both steps, and a Markov random field formulation that incorporates tissue coherency to compute the nonrigid registration. To fix significantly distorted regions that are misshaped or much smaller than the control grids, we train a context-aggregation network to segment and warp them to their corresponding regions with thin plate spline. We have shown that our method generates results comparable to an expert neuroscientist and is significantly better than reconstruction-first approaches.
Materials
Source Code and Data (The code is matlab and can be used on Linux machines. We tested with Windows machine but due to large data size of Allen Atlas and experimental brains, and MATLAB windows version memory limit, we got "out of memory" errors. We recommend a >2GB RAM for running the program.
Applications
Our work was used to map all the experimental brain stacks to the Allen Mouse Brain Atlas in paper "Anatomically Defined and Functionally Distinct Dorsal Raphe Serotonin Sub-systems" that was published on Cell.
Our work is currently under use in another project to map the more posterior brain to the Allen Mouse Brain Atlas.