RayOoooo / BitDepth_NucSeg

This repository contains the fully annotated nuclei segmentation masks for 8-bit and 16-bits fluorescence Dapi stained microscopic images

BitDepth_NucSeg

This repository contains a fully annotated nuclei segmentation dataset of 8-bit and 16-bits fluorescence Dapi stained microscopic images from 5 human organs at a fixed size of 512x512 pixels.

Citation

A paper describing our dataset and methodology to compare the instance segmentation performance of 8-bit vs. 16-bit fluorescence Dapi stained microscopic images is accepted to the Diagnostics Journal. BibTex entry:

@Article{diagnostics11060967,
AUTHOR = {Mahbod, Amirreza and Schaefer, Gerald and Löw, Christine and Dorffner, Georg and Ecker, Rupert and Ellinger, Isabella},
TITLE = {Investigating the Impact of the Bit Depth of Fluorescence-Stained Images on the Performance of Deep Learning-Based Nuclei Instance Segmentation},
JOURNAL = {Diagnostics},
VOLUME = {11},
YEAR = {2021},
NUMBER = {6},
ARTICLE-NUMBER = {967},
URL = {https://www.mdpi.com/2075-4418/11/6/967},
ISSN = {2075-4418},
DOI = {10.3390/diagnostics11060967}
}

paper link: https://www.mdpi.com/2075-4418/11/6/967

Acknowledgements

This work was supported by the Austrian Research Promotion Agency (FFG), No. 872636.