CytoDArk0
收藏资源简介:
CytoDArk0是由伦敦南岸大学团队创建的一个用于脑细胞实例分割和分类的新型数据集,主要包含Nissl染色的大脑皮质、小脑和海马体的图像。该数据集涵盖了来自鲸偶蹄目和灵长目动物的样本,旨在支持脑细胞结构研究的自动化和可重复性。数据集的创建过程包括初始的机器学习方法和逐步的深度学习方法,最终通过CISCA框架进行处理。CytoDArk0的应用领域包括神经退行性疾病和神经炎症疾病的研究,旨在通过精确量化脑细胞的类型、密度和空间排列,揭示疾病的发病机制和治疗方法。
CytoDArk0 is a novel dataset developed by the team at London South Bank University for instance segmentation and classification of brain cells. It mainly comprises Nissl-stained images of the cerebral cortex, cerebellum and hippocampus. The dataset encompasses samples from cetartiodactyls and primates, with the goal of supporting automation and reproducibility in brain cell structure research. The creation of this dataset involved initial machine learning approaches and incremental deep learning methods, and was ultimately processed using the CISCA framework. Applications of CytoDArk0 cover research into neurodegenerative and neuroinflammatory diseases, where it aims to reveal disease pathogenesis and therapeutic strategies via precise quantification of brain cell types, density and spatial organization.

- 1CISCA and CytoDArk0: a Cell Instance Segmentation and Classification method for histo(patho)logical image Analyses and a new, open, Nissl-stained dataset for brain cytoarchitecture studies伦敦南岸大学 · 2024年



