Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

An annotated high-content fluorescence microscopy dataset with Hoechst 33342-stained nuclei and manually labelled outlines

Arvidsson, Malou ; Rashed, Salma Kazemi LU and Aits, Sonja LU orcid (2023) In Data in Brief 46.
Abstract

Automated detection of cell nuclei in fluorescence microscopy images is a key task in bioimage analysis. It is essential for most types of microscopy-based high-throughput drug and genomic screening and is often required in smaller scale experiments as well. To develop and evaluate algorithms and neural networks that perform instance or semantic segmentation for detecting nuclei, high quality annotated data is essential. Here we present a benchmarking dataset of fluorescence microscopy images with Hoechst 33342-stained nuclei together with annotations of nuclei, nuclear fragments and micronuclei. Images were randomly selected from an RNA interference screen with a modified U2OS osteosarcoma cell line, acquired on a Thermo Fischer CX7... (More)

Automated detection of cell nuclei in fluorescence microscopy images is a key task in bioimage analysis. It is essential for most types of microscopy-based high-throughput drug and genomic screening and is often required in smaller scale experiments as well. To develop and evaluate algorithms and neural networks that perform instance or semantic segmentation for detecting nuclei, high quality annotated data is essential. Here we present a benchmarking dataset of fluorescence microscopy images with Hoechst 33342-stained nuclei together with annotations of nuclei, nuclear fragments and micronuclei. Images were randomly selected from an RNA interference screen with a modified U2OS osteosarcoma cell line, acquired on a Thermo Fischer CX7 high-content imaging system at 20x magnification. Labelling was performed by a single annotator and reviewed by a biomedical expert. The dataset, called Aitslab-bioimaging1, contains 50 images showing over 2000 labelled nuclear objects in total, which is sufficiently large to train well-performing neural networks for instance or semantic segmentation. The dataset is split into training, development and test set for user convenience.

(Less)
Please use this url to cite or link to this publication:
author
; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Biomedical image analysis, Computer vision, Deep learning training and evaluation, Fluorescence microscopy, High-content screening, Instance segmentation
in
Data in Brief
volume
46
article number
108769
publisher
Elsevier
external identifiers
  • scopus:85143768599
  • pmid:36506804
ISSN
2352-3409
DOI
10.1016/j.dib.2022.108769
project
Lysosomes in cell death - from molecular mechanisms to new treatment strategies
Understanding and therapeutic targeting of lysosome-dependent cell death
Lund University AI Research
Analysing large-scale microscopy datasets with AI-based approaches
Environmental toxins as cause of lysosomal damage-induced cell death in animals and humans
Using artificial intelligence and advanced computational tools to identify biological pathways, disease mechanisms and therapeutic opportunities
language
English
LU publication?
yes
id
4451e64a-c796-461f-9d4d-35fecfe82b82
date added to LUP
2022-12-15 08:38:09
date last changed
2024-04-18 06:33:03
@article{4451e64a-c796-461f-9d4d-35fecfe82b82,
  abstract     = {{<p>Automated detection of cell nuclei in fluorescence microscopy images is a key task in bioimage analysis. It is essential for most types of microscopy-based high-throughput drug and genomic screening and is often required in smaller scale experiments as well. To develop and evaluate algorithms and neural networks that perform instance or semantic segmentation for detecting nuclei, high quality annotated data is essential. Here we present a benchmarking dataset of fluorescence microscopy images with Hoechst 33342-stained nuclei together with annotations of nuclei, nuclear fragments and micronuclei. Images were randomly selected from an RNA interference screen with a modified U2OS osteosarcoma cell line, acquired on a Thermo Fischer CX7 high-content imaging system at 20x magnification. Labelling was performed by a single annotator and reviewed by a biomedical expert. The dataset, called Aitslab-bioimaging1, contains 50 images showing over 2000 labelled nuclear objects in total, which is sufficiently large to train well-performing neural networks for instance or semantic segmentation. The dataset is split into training, development and test set for user convenience.</p>}},
  author       = {{Arvidsson, Malou and Rashed, Salma Kazemi and Aits, Sonja}},
  issn         = {{2352-3409}},
  keywords     = {{Biomedical image analysis; Computer vision; Deep learning training and evaluation; Fluorescence microscopy; High-content screening; Instance segmentation}},
  language     = {{eng}},
  publisher    = {{Elsevier}},
  series       = {{Data in Brief}},
  title        = {{An annotated high-content fluorescence microscopy dataset with Hoechst 33342-stained nuclei and manually labelled outlines}},
  url          = {{http://dx.doi.org/10.1016/j.dib.2022.108769}},
  doi          = {{10.1016/j.dib.2022.108769}},
  volume       = {{46}},
  year         = {{2023}},
}