What it is
Drosophila melanogaster Kc167 cells, imaged in two channels: DNA, which shows the nuclei, and actin, which shows the cell body. Cells are about 25 px across, with about 80 cells per image on average.
| Fact | Value |
|---|---|
| Accession | BBBC007, version 1 |
| Cells | Drosophila melanogaster Kc167 |
| Channels | DNA (nuclei), actin (cell body), separate files |
| Format | 8-bit grayscale TIFF, roughly 512 x 512 px |
| Ground truth | hand-drawn outlines of nuclei and cells, 8-bit monochrome TIFF |
| License | CC0 1.0 |
Browse it
The DataTorch copy has the outlines converted to polygons, so you can see every nucleus and cell drawn on the image before downloading anything.
Dataset · datatorch/cell-segmentation-bbbc007 · 32 files
Cell Segmentation (BBBC007)
Two-channel cell segmentation dataset of Drosophila melanogaster Kc167 cells (BBBC007v1, Broad Bioimage Benchmark Collection). 16 fields of view, each as a DNA image and an actin image (32 grayscale 8-bit images, 400 to 512 px), with 1,264 hand-outlined nuclei and 1,221 hand-outlined cells converted from the official outline rasters to polygon annotations. The benchmark task is finding the boundaries between touching cells, using the nuclei as seeds. License: CC0 1.0 (public domain). Source: https://bbbc.broadinstitute.org/BBBC007. Please cite: Jones, Carpenter & Golland, Voronoi-based segmentation of cells on image manifolds, Proc. ICCV Workshop on Computer Vision for Biomedical Image Applications (CVBIA), 2005; and Ljosa, Sokolnicki & Carpenter, Annotated high-throughput microscopy image sets for validation, Nature Methods 9, 637 (2012).
Open →
| In the hosted copy | Value |
|---|---|
| Fields of view | 16 (32 images, 400 to 512 px) |
| Nuclei | 1,264 |
| Cells | 1,221 |
What people use it for
Seeded segmentation: find the nuclei first, then grow each cell outward until it meets its neighbours. The hand outlines show where that boundary really is.
Source and citation
Source: https://bbbc.broadinstitute.org/BBBC007 (read 2026-10-22). Cite image set BBBC007v1 (Jones et al., Proc. ICCV Workshop on Computer Vision for Biomedical Image Applications, 2005), available from the Broad Bioimage Benchmark Collection [Ljosa et al., Nature Methods, 2012].
Outline your own cells
Draw, correct and review polygons with your team in the browser.
Frequently asked questions
What does BBBC007 contain?
Two-channel images of Drosophila Kc167 cells (DNA and actin, as separate 8-bit grayscale TIFFs) with hand-drawn outlines of nuclei and cells, one cell outline per nucleus.
How many objects are outlined in BBBC007?
The copy hosted on DataTorch has 16 fields of view (32 images) with 1,264 nuclei and 1,221 cells as polygons, converted from the official outline rasters.
What is the license of BBBC007?
CC0 1.0. Anne Carpenter waived copyright and related rights to the images and ground truth.
Open the hosted copy on DataTorch.
Browse the files in your browser, then download the dataset with a free account.
