遇见数据集

H&E- and Nissl-Stained Histopathological Rabbit Brain Image Dataset for Photothrombotic Lesion Classification

收藏
Zenodo2026-08-15 更新2026-08-20 收录
官方服务:

资源简介:

This dataset contains histopathological image patches obtained from rabbit brain tissues following photothrombotic ischemic stroke induction. It was developed for multiclass classification of photothrombotic lesions, regions containing necrotic neuronal cells, and contralesional normal brain tissue. Brain tissue sections were stained using Hematoxylin and Eosin (H&E) or Cresyl Violet (Nissl) staining and digitized using a Zeiss Axio Scan.Z1 slide scanner. Image regions were manually identified and extracted from three tissue categories: normal tissue, necrotic regions, and photothrombotic lesions. The source whole-slide images (WSIs) were obtained from the photothrombotic rabbit brain specimens described by Kim et al.: Y. Kim, Y. B. Lee, S. K. Bae, S. S. Oh, and J.-r. Choi, “Development of a photochemical thrombosis investigation system to obtain a rabbit ischemic stroke model,” Scientific Reports, vol. 11, article 5787, 2021.https://doi.org/10.1038/s41598-021-85348-6 The image patches included in this dataset were derived from those source WSIs. Each staining-specific dataset contains 2,175 image patches: 725 photothrombotic lesion images 725 necrotic-region images 725 normal tissue images The complete dataset contains 4,350 histopathological images across the H&E and Nissl staining modalities. The images are organized into class-specific directories and stored in BMP format. The dataset was used and evaluated in the following article. Users of this dataset are requested to cite this article: Choi, J.-r., Jeon, M., Choi, S. W., and Oh, T., “A comparative study on deep learning architectures for a classification of photothrombotic damaged regions in histopathological rabbit brain images,” Biomedical Signal Processing and Control, vol. 111, article 108354, 2026.https://doi.org/10.1016/j.bspc.2025.108354 The dataset supports research on histopathological image classification, ischemic stroke analysis, photothrombotic lesion assessment, deep learning model benchmarking, and preclinical brain tissue analysis. Animal experiments were approved by the Animal Experiment Ethics Committee of the Daegu-Gyeongbuk Medical Innovation Foundation under approval number DGMIF-20061702-00. ContactFor questions regarding the dataset, please contact Taegeun Oh, Department of Electronic Engineering, Dong Seoul University, at tgoh@du.ac.kr Jong-ryul Choi, Medical Device Development Center, Daegu-Gyeongbuk Medical Innovation Foundation (K-MEDI hub), at jongryul32@kmedihub.re.kr

提供机构:
Zenodo
创建时间:
2026-08-04
二维码
社区交流群
二维码
科研交流群
商业服务