One of the principal challenges in counting or segmenting cells or cell nuclei is dealing with clustered objects. To help assess algorithms' performance in this regard, synthetic 3D image sets of HL60 cell line are provided consisting of four subsets with increasing degree of clustering. Each subset is also provided in two diferent levels of quality: high SNR and low SNR.Ground truth is available as well.
One of the principal challenges in counting or segmenting cells or cell nuclei is dealing with clustered objects such as in tissues. To help assess algorithms' performance in this regard, synthetic 3D image sets of human colon tissue are provided in two diferent levels of quality: high SNR and low SNR. Ground truth is available as well.
A fully annotated dataset of Two-Photon Laser Scanning Microscopy (2PLSM) images of three types of dendritic spines. We make a standard dendritic analysis dataset publicly available including raw data, manual annotations, and manual labels. Manual labels and annotation (segmentations) are provided by a neuroscience expert. The dataset has been used in Ghani2016 to perform dendritic spine analysis.
Fluorescence microscopy cannot be used to image human embryos to determine embryo viability for in vitro fertilization because the introduction of exogenous fluorescent dyes is considered a toxic procedure. As a result, embryo viability has been measured primarily using differential interference contrast (DIC). A human can readily segment the embryo (and, to some extent, individual cells) in a DIC image, but automatic segmentation remains a challenge due to the cosine-dependent shading inherent in DIC images. There are 15 images.
Five different samples of Drosophila melanogaster Kc167 cells were stained with Hoechst 33342, a DNA stain. The last sample (labeled nodsRNA) is of wild-type cells. Each of the other four samples (labeled 48, 340, Anillin, and mad2) has a different gene knocked down by RNAi. The sample preparation is described in more detail by [Carpenter et al.
In terms of image analysis, this article includes an analysis of 2D myofibre labeled with a dye to identify myofibre borders. Keefe, A.C. et al. Muscle stem cells contribute to myofibres in sedentary adult mice. Nat. Commun. 6:7087 doi: 10.1038/ncomms8087 (2015). An open source version based on python is available at: https://github.com/stevendflygare/muscleQNT
Two-channel images (brightfield and fluorescent DNA stain) of HeLa cells in culture. Cell and nucleus outlines are manually labelled. Very suitable for cell segmentation benchmark tests. Published here: Van Valen, David A., et al. "Deep Learning Automates the Quantitative Analysis of Individual Cells in Live-Cell Imaging Experiments." PLoS Comput Biol 12.11 (2016): e1005177. http://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1…