Image segmentation and registration of carp brain tissue slices oriented to brain atlas construction

Brain atlas is an important tool to provide information on the location and function of brain motor nuclei, and the segmentation and registration of brain tissue slice images is the necessary basis for building brain atlases.

In order to construct the carp brain atlas, the segmentation and the registration of carp brain tissue images were studied in this study. In this study, carp brain tissue sections were prepared by the method of pre-fixation of brain tissue specimens and paraffin tissue sections with HE staining techniques, and a multi-threshold image segmentation algorithm based on HSI color space was selected for image segmentation according to the image characteristics of carp brain tissue sections. In the aspect of image registration, a method of selecting feature points for registration based on the morphological and structural characteristics of different brain regions of carp was used for image registration of brain tissue slices.

The results showed that the image segmentation and registration algorithm used in this study can meet the experimental requirements and conform to the structural characteristics of carp brain in spatial position.

The traditional brain mapping is drawn manually by humans, which is time-consuming and labor-intensive, and is limited to two dimensions. Image segmentation technology can automatically identify the contours of brain regions, saving manpower and promoting three-dimensional reconstruction and digital research.

This study laid a foundation for the construction of stereotaxic map of carp brain.