Mouse brain histology reconstruction based on piecewise registration and correction

Histology acquisition often introduces tissue distortions that make 3D reconstruction difficult. In this paper, we propose a method to detect tissue distortion within histology slices by running a piecewise registration scheme. It takes into account the information of consecutive slices in a neighborhood. The method runs iteratively to borrow undistorted tissue information from adjacent slices to correct distortions. A smooth mouse brain volume is then reconstructed by co-registering the corrected slices together.