Accession number BBBC024 · Version 1
|3D image||3D foreground|
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.
Four subsets (each in high and low SNR variant) of 30 images each are provided. Each image contains 20 HL60 cell nuclei, but the nuclei cluster with different probabilities (0%, 25%, 50%, and 75%) in the four subsets. The dataset was generated using the virtual microscope imitating the microscope Zeiss S100 (objective Zeiss 63x/1.40 Oil DIC) attached to confocal unit Atto CARV and CCD camera Micromax 1300-YHS.
Each image contains exactly 20 masks; this is the ground truth for counting. Ground truth for foreground/background segmentation are available as labeled 16bit grayscale images:
|clustering probability||SNR||download foreground|
"We used image set BBBC024vl [Svoboda David, Kozubkek Michal, Stejskal Stanislav. Generation of Digital Phantoms of Cell Nuclei and Simulation of Image Formation in 3D Image Cytometry. Cytometry Part A, John Wiley & Sons, Inc., 75A, 6, pp. 494-509, 16 pages. ISSN 1552-4922. 2009.] from the Broad Bioimage Benchmark Collection."
The images and ground truth are licensed under a Creative Commons Attribution 3.0 Unported License by David Svoboda.