Bbbc.broadinstitute.org (12)#
3D HL60 Cell line (synthetic data)#
David Svoboda, Michal Kozubkek, Stanislav Stejskal
Published 2009-06-01
Licensed CC-BY-3.0
One of the principal challenges in counting or segmenting nuclei is dealing with clustered nuclei. To help assess algorithms performance in this regard, this synthetic image set consists of four subsets with increasing degree of clustering. Each subset is also provided in two diferent levels of quality: high SNR and low SNR.
Tags: Ai-Ready
Content type: Data
Annotated high-throughput microscopy image sets for validation#
Vebjorn Ljosa, Katherine L Sokolnicki, Anne E Carpenter
Broad Bioimage Benchmark Collection (BBBC)
Content type: Collection, Data
Chinese Hamster Ovary Cells#
Krisztian Koos, József Molnár, Lóránd Kelemen, Gábor Tamás, Peter Horvath
Published 2016-07-29
Licensed CC-BY-3.0
The image set consists of 60 Differential Interference Contrast (DIC) images of Chinese Hamster Ovary (CHO) cells. The images are taken on an Olympus Cell-R microscope with a 20x lens at the time when the cell initiated their attachment to the bottom of the dish.
Tags: Ai-Ready
Content type: Data
Drosophila Kc167 cells#
Vebjorn Ljosa, Katherine L. Sokolnicki, Anne E. Carpenter
Published 2012-06-28
Licensed CC0-1.0
Drosophila melanogaster Kc167 cells were stained for DNA (to label nuclei) and actin (a cytoskeletal protein, to show the cell body). Automatic cytometry requires that cells be segmented, i.e., that the pixels belonging to each cell be identified. Because segmenting nuclei and distinguishing foreground from background is comparatively easy for these images, the focus here is on finding the boundaries between adjacent cells.
Tags: Ai-Ready
Content type: Data
Human HT29 colon-cancer cells#
Vebjorn Ljosa, Katherine L. Sokolnicki, Anne E. Carpenter
Published 2012-06-28
Licensed CC-BY-NC-SA-3.0
These images are of human HT29 colon cancer cells, a cell line that has been widely used for the study of many normal and neoplastic processes. A set of about 43,000 such images was used by Moffat et al. (Cell, 2006) to screen for mitotic regulators. The analysis followed the common pattern of identifying and counting cells with a phenotype of interest (in this case, cells that were in mitosis), then normalizing the count by dividing by the total number of cells. Such experiments present two image analysis problems. First, identifying the cells that have the phenotype of interest requires that the nuclei and cells be segmented. Second, normalizing requires an accurate cell count.
Tags: Ai-Ready
Content type: Data
Human Hepatocyte and Murine Fibroblast cells Co-culture experiment#
David J. Logan, Jing Shan, Sangeeta N. Bhatia, Anne E. Carpenter
Published 2016-03-01
Licensed CC-BY-3.0
This 384-well plate has images of co-cultured hepatocytes and fibroblasts. Every other well is populated (A01, A03, …, C01, C03, …) such that 96 wells comprise the data. Each well has 9 sites and thus 9 images associated, totaling 864 images.
Tags: Ai-Ready
Content type: Data
Human U2OS cells (out of focus)#
Vebjorn Ljosa, Katherine L. Sokolnicki, Anne E. Carpenter
Published 2012-06-28
Licensed CC0-1.0
Since robust foreground/background separation and segmentation of cellular objects (i.e., identification of which pixels below to which objects) strongly depends on image quality, focus artifacts are detrimental to data quality. This image set provides examples of in- and out-of-focus HCS images which can be used for validation of focus metrics.
Tags: Ai-Ready
Content type: Data
Mouse embryo blastocyst cells#
Vebjorn Ljosa, Katherine L. Sokolnicki, Anne E. Carpenter
Published 2012-06-28
Licensed CC0-1.0
Segmenting nuclei in 3D images can be challenging especially when nuclei are clustered not only in XY plane but also in XZ and YZ planes. Manually annotated ground truth provides a reference for image analysis software testing purposes. These images of mouse embryo blastocyst cells also have changing nuclei intensity in Z plane which makes finding the right threshold for successful segmentation a difficult task. This image set also contains GAPDH transcripts that can be quantified in each cell.
Tags: Ai-Ready
Content type: Data
Nuclei of U2OS cells in a chemical screen#
Vebjorn Ljosa, Katherine L. Sokolnicki, Anne E. Carpenter
Published 2012-06-28
Licensed CC0-1.0
This image set is part of a high-throughput chemical screen on U2OS cells, with examples of 200 bioactive compounds. The effect of the treatments was originally imaged using the Cell Painting assay (fluorescence microscopy). This data set only includes the DNA channel of a single field of view per compound. These images present a variety of nuclear phenotypes, representative of high-throughput chemical perturbations. The main use of this data set is the study of segmentation algorithms that can separate individual nucleus instances in an accurate way, regardless of their shape and cell density. The collection has around 23,000 single nuclei manually annotated to establish a ground truth collection for segmentation evaluation.
Tags: Ai-Ready
Content type: Data
Nuclei of mouse embryonic cells#
Vebjorn Ljosa, Katherine L. Sokolnicki, Anne E. Carpenter
Published 2012-06-28
Licensed CC0-1.0
Cell dynamics during the early mouse embryogenesis change spatiotemporally. For understanding the mechanism of this developmental process, imaging cell dynamics by live-cell imaging of fluorescently labeled nuclei and performing nuclei segmentation of these images by image processing are essential. This dataset contains the fluorescence images and Ground Truth used when performing nuclei segmentation using deep learning. Fluorescence images are time-series images from fertilization to blastocyst formation. Ground Truth is supervised data of the cell nuclear region.
Tags: Ai-Ready
Content type: Data
Simulated HL60 cells (from the Cell Tracking Challenge)#
Vebjorn Ljosa, Katherine L. Sokolnicki, Anne E. Carpenter
Published 2012-06-28
Licensed CC0-1.0
These are synthetic images from the Cell Tracking Challenge. The images depict simulated nuclei of HL60 cells stained with Hoescht (training datasets). These synthetic images of HL60 cells provide an opportunity to test image analysis software by comparing segmentation results to the available ground truth for each time point. The number of clustered nuclei increases with time adding more complexity to the problem. This time-laps dataset can be used for simple segmentation or for nuclei tracking.
Tags: Ai-Ready
Content type: Data
Synthetic cells#
Vebjorn Ljosa, Katherine L. Sokolnicki, Anne E. Carpenter
Published 2012-06-28
Licensed CC-BY-NC-SA-3.0
One of the principal challenges in counting or segmenting nuclei is dealing with clustered nuclei. To help assess algorithms performance in this regard, this synthetic image set consists of five subsets with increasing degree of clustering.
Tags: Ai-Ready
Content type: Data
https://bbbc.broadinstitute.org/BBBC004