Accession number BBBC042 · Version 1
Description of the biological application
The morphology of astrocytes is highly heterogeneous and counting them is currently a laborious task. Development of fully automated counting methods for astrocytes from immunohistological images without any user intervention is one of the major challenges.
There are 1200 images. The images were saved from scanned slides. Slides were scanned using the 3DHISTECH Scanner (3DHISTECH Ltd, Budapest, Hungary). The image size is 990 × 708 pixels. Images are available in 8-bit TIF format.
The images contain GFAP staining in different rat brain regions. These images were used to train Deep convolutional neural network (DCNN). Corresponding text files contain coordinates of labeled cells to train DCNN. 15000 cells were labeled.
For more information
Please contact Ilida Suleymanova regarding this dataset.
Published results using this image set
Suleymanova I, Balassa T, Tripathi S, et al. A deep convolutional neural network approach for astrocyte detection. Scientific Reports. 2018;8:12878. doi:10.1038/s41598-018-31284-x. PMCID: PMC6110828.
"We used image set BBBC042v1 Suleymanova I. et al., available from the Broad Bioimage Benchmark Collection [Ljosa et al., Nature Methods, 2012]."
Copyright: CC0. To the extent possible under law, Ilida Suleymanova has waived all copyright and related or neighboring rights to BBBC042v1.