High-resolution coronal sections of mouse brain with neuron segmentation and feature extraction (v1)

This dataset comprises ten high-resolution images of coronal sections of a mouse brain located in antero-posterior hippocampal region and corresponding delineation of anatomical regions of interest.  It also includes the instance segmentation of neurons across all sections, along with extracted neuronal features. Sections were stained by immunohistochemistry using the neural Nuclei (NeuN) antibody staining neurons. The sections were digitized at 20x in bright field (~3-5GB / section) by a virtual scanner, Axioscan.Z1 (Zeiss). An expert in neurobiology has manually segmented the main anatomical brain regions using the Allen Atlas as a reference.
Neuron segmentation maps were computed with a deep learning algorithm specifically designed for neuron segmentation ([Wu et al., 2022](https://doi.org/10.1016/j.compbiomed.2022.106180)) and optimized with high-performance computing, allowing the process of these multi-gigabyte images. As a result, approximately 900k neurons were segmented. Fifty neuron characteristics, including localisation and morphological features (area, circularity, etc.), were computed on all neurons and provided in the dataset in a structured text format. 
This dataset supports a range of analyses focused on neuronal populations, including counting, morphological profiling and spatial distribution studies, which are of major interest in both physiological and pathophysiological conditions. Furthermore, it provides a valuable resource for investigating the innovative approach of cell-graph representation of cell populations in entire brain sections to perform multiscale analyses (cell population description, automated anatomical segmentation).