Hyperspectral Histological Images for Diagnosis of Human Glioblastoma (HistologyHSI-GB)

Hyperspectral imaging technology combines the main features of imaging and spectroscopy, allowing simultaneous analysis of the morphological features and chemical composition of samples in a non-contact way. The information provided by hyperspectral imaging can be used to detect patterns, cells, or biomarkers to identify diseases. This dataset is composed of hyperspectral histological images of glioblastoma brain tumors, which can be used by the research community to analyze the potential of hyperspectral images for diverse histological applications. The dataset contains 469 annotated hyperspectral images corresponding to 13 histological slides (1 per patient). The hyperspectral images are captured with a 20× magnification in the spectral range from 400 – 1000 nm. Each hyperspectral image has been annotated based on the diagnosis of the slides by skilled histopathologists into the categories of glioblastoma and no tumor according to the World Health Organization classification of tumors of the nervous system.

高光谱成像（hyperspectral imaging）技术兼具成像与光谱学的核心特性，可实现非接触式同步分析样本的形态学特征与化学成分。该技术获取的信息可用于检测特征模式、细胞或生物标志物，进而实现疾病识别。本数据集由脑胶质母细胞瘤（glioblastoma）的高光谱组织学图像构成，可供科研界探究高光谱成像在各类组织学应用场景中的应用潜力。数据集包含469张已标注高光谱图像，对应13张组织切片（每例患者对应1张切片）。这批高光谱图像以20倍放大倍率采集，光谱覆盖范围为400 nm至1000 nm。所有高光谱图像均由经验丰富的组织病理学家，参照世界卫生组织（World Health Organization, WHO）神经系统肿瘤分类标准，结合切片诊断结果标注为胶质母细胞瘤与无肿瘤两类。