Electron Microscopy 3D Segmentation

# Context

The dataset available for download on this webpage represents a 5x5x5µm section taken from the CA1 hippocampus region of the brain, corresponding to a 1065x2048x1536 volume. The resolution of each voxel is approximately 5x5x5nm.

# Content

Two image datasets in 3D of Electron Microscopy data with accompanying labels. The data is provided as multipage TIF files that can be loaded in Fiji, R, KNIME, or Python

# Acknowledgements

The dataset was copied from http://cvlab.epfl.ch/data/em directly and only placed here to utilize the Kaggle's kernel and forum capabilities. Please acknowledge the CV group dataset for publication or any other uses

## Data Citations

- A. Lucchi Y. Li and P. Fua, Learning for Structured Prediction Using Approximate Subgradient Descent with Working Sets, Conference on Computer Vision and Pattern Recognition, 2013.
- A. Lucchi, K.Smith, R. Achanta, G. Knott, P. Fua, Supervoxel-Based
Segmentation of Mitochondria in EM Image Stacks with Learned Shape
Features, IEEE Transactions on Medical Imaging, Vol. 30, Nr. 11,
October 2011.

# Challenges
- How accurately can the segmentation be performed with neural networks?
- Is 3D more accurate than 2D for segmentation?
- How can mistakes critical to structure or connectivity be penalized more heavily, how would a standard ROC penalize them?

{'# Context': '本网页提供的可供下载的数据集，系从大脑海马体CA1区域的5x5x5µm切片获取，对应于1065x2048x1536的体积。每个体素的分辨率约为5x5x5nm。', '# Content': '包含两个3D电子显微镜数据图像数据集，并附有相应的标签。数据以多页TIF文件形式提供，可加载于Fiji、R、KNIME或Python中。', '# Acknowledgements': '本数据集直接复制自http://cvlab.epfl.ch/data/em，仅在此处放置以利用Kaggle的内核和论坛功能。请在出版物或其他用途中提及CV组数据集。', '# Data Citations': {'-': 'A. Lucchi, K. Smith, R. Achanta, G. Knott, P. Fua. 基于超像素的线粒体在电子显微镜图像序列中的分割，利用学习到的形状特征，IEEE医学成像 Transactions，第30卷，第11期，2011年10月。'}, '# Challenges': {'-': '如何对结构或连接中的关键错误进行更重的惩罚，标准ROC曲线如何惩罚它们？'}}