Nuclei segmentation of fluorescence microscopy images using convolutional neural networks

Chichen Fu, David Joon Ho, Shuo Han, Paul Salama, Kenneth Dunn, Edward J. Delp

Research output: Chapter in Book/Report/Conference proceedingConference contribution

7 Citations (Scopus)

Abstract

Fluorescence microscopy has emerged as a powerful tool for studying cell biology because it enables the acquisition of 3D image volumes deeper into tissue and the imaging of complex subcellular structures. Quantitative analysis of these structures, which is needed to characterize the structure and constitution of tissue volumes, is facilitated by nuclei segmentation. However, manual segmentation is a laborious and intractable process due to the size and complexity of the data. In this paper, we describe a nuclei segmentation method using a deep convolutional neural network, data augmentation to generate training images of different shapes and contrasts, a refinement process combining segmentation results of horizontal, frontal, and sagittal planes in a volume, and a watershed technique to count the number of nuclei. Our results indicate that compared to 3D ground truth data, our method is able to successfully segment and count 3D nuclei.

Original languageEnglish (US)
Title of host publication2017 IEEE 14th International Symposium on Biomedical Imaging, ISBI 2017
PublisherIEEE Computer Society
Pages704-708
Number of pages5
ISBN (Electronic)9781509011711
DOIs
StatePublished - Jun 15 2017
Event14th IEEE International Symposium on Biomedical Imaging, ISBI 2017 - Melbourne, Australia
Duration: Apr 18 2017Apr 21 2017

Other

Other14th IEEE International Symposium on Biomedical Imaging, ISBI 2017
CountryAustralia
CityMelbourne
Period4/18/174/21/17

Fingerprint

Fluorescence microscopy
Fluorescence Microscopy
Cytology
Tissue
Neural networks
Constitution and Bylaws
Watersheds
Cell Biology
Imaging techniques
Chemical analysis

Keywords

  • 3D ground truth
  • Convolutional neural network
  • Data augmentation
  • Fluorescence microscopy
  • Nuclei segmentation

ASJC Scopus subject areas

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging

Cite this

Fu, C., Ho, D. J., Han, S., Salama, P., Dunn, K., & Delp, E. J. (2017). Nuclei segmentation of fluorescence microscopy images using convolutional neural networks. In 2017 IEEE 14th International Symposium on Biomedical Imaging, ISBI 2017 (pp. 704-708). [7950617] IEEE Computer Society. https://doi.org/10.1109/ISBI.2017.7950617

Nuclei segmentation of fluorescence microscopy images using convolutional neural networks. / Fu, Chichen; Ho, David Joon; Han, Shuo; Salama, Paul; Dunn, Kenneth; Delp, Edward J.

2017 IEEE 14th International Symposium on Biomedical Imaging, ISBI 2017. IEEE Computer Society, 2017. p. 704-708 7950617.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Fu, C, Ho, DJ, Han, S, Salama, P, Dunn, K & Delp, EJ 2017, Nuclei segmentation of fluorescence microscopy images using convolutional neural networks. in 2017 IEEE 14th International Symposium on Biomedical Imaging, ISBI 2017., 7950617, IEEE Computer Society, pp. 704-708, 14th IEEE International Symposium on Biomedical Imaging, ISBI 2017, Melbourne, Australia, 4/18/17. https://doi.org/10.1109/ISBI.2017.7950617
Fu C, Ho DJ, Han S, Salama P, Dunn K, Delp EJ. Nuclei segmentation of fluorescence microscopy images using convolutional neural networks. In 2017 IEEE 14th International Symposium on Biomedical Imaging, ISBI 2017. IEEE Computer Society. 2017. p. 704-708. 7950617 https://doi.org/10.1109/ISBI.2017.7950617
Fu, Chichen ; Ho, David Joon ; Han, Shuo ; Salama, Paul ; Dunn, Kenneth ; Delp, Edward J. / Nuclei segmentation of fluorescence microscopy images using convolutional neural networks. 2017 IEEE 14th International Symposium on Biomedical Imaging, ISBI 2017. IEEE Computer Society, 2017. pp. 704-708
@inproceedings{51fec6378d5e466a97ed92ae0e327247,
title = "Nuclei segmentation of fluorescence microscopy images using convolutional neural networks",
abstract = "Fluorescence microscopy has emerged as a powerful tool for studying cell biology because it enables the acquisition of 3D image volumes deeper into tissue and the imaging of complex subcellular structures. Quantitative analysis of these structures, which is needed to characterize the structure and constitution of tissue volumes, is facilitated by nuclei segmentation. However, manual segmentation is a laborious and intractable process due to the size and complexity of the data. In this paper, we describe a nuclei segmentation method using a deep convolutional neural network, data augmentation to generate training images of different shapes and contrasts, a refinement process combining segmentation results of horizontal, frontal, and sagittal planes in a volume, and a watershed technique to count the number of nuclei. Our results indicate that compared to 3D ground truth data, our method is able to successfully segment and count 3D nuclei.",
keywords = "3D ground truth, Convolutional neural network, Data augmentation, Fluorescence microscopy, Nuclei segmentation",
author = "Chichen Fu and Ho, {David Joon} and Shuo Han and Paul Salama and Kenneth Dunn and Delp, {Edward J.}",
year = "2017",
month = "6",
day = "15",
doi = "10.1109/ISBI.2017.7950617",
language = "English (US)",
pages = "704--708",
booktitle = "2017 IEEE 14th International Symposium on Biomedical Imaging, ISBI 2017",
publisher = "IEEE Computer Society",

}

TY - GEN

T1 - Nuclei segmentation of fluorescence microscopy images using convolutional neural networks

AU - Fu, Chichen

AU - Ho, David Joon

AU - Han, Shuo

AU - Salama, Paul

AU - Dunn, Kenneth

AU - Delp, Edward J.

PY - 2017/6/15

Y1 - 2017/6/15

N2 - Fluorescence microscopy has emerged as a powerful tool for studying cell biology because it enables the acquisition of 3D image volumes deeper into tissue and the imaging of complex subcellular structures. Quantitative analysis of these structures, which is needed to characterize the structure and constitution of tissue volumes, is facilitated by nuclei segmentation. However, manual segmentation is a laborious and intractable process due to the size and complexity of the data. In this paper, we describe a nuclei segmentation method using a deep convolutional neural network, data augmentation to generate training images of different shapes and contrasts, a refinement process combining segmentation results of horizontal, frontal, and sagittal planes in a volume, and a watershed technique to count the number of nuclei. Our results indicate that compared to 3D ground truth data, our method is able to successfully segment and count 3D nuclei.

AB - Fluorescence microscopy has emerged as a powerful tool for studying cell biology because it enables the acquisition of 3D image volumes deeper into tissue and the imaging of complex subcellular structures. Quantitative analysis of these structures, which is needed to characterize the structure and constitution of tissue volumes, is facilitated by nuclei segmentation. However, manual segmentation is a laborious and intractable process due to the size and complexity of the data. In this paper, we describe a nuclei segmentation method using a deep convolutional neural network, data augmentation to generate training images of different shapes and contrasts, a refinement process combining segmentation results of horizontal, frontal, and sagittal planes in a volume, and a watershed technique to count the number of nuclei. Our results indicate that compared to 3D ground truth data, our method is able to successfully segment and count 3D nuclei.

KW - 3D ground truth

KW - Convolutional neural network

KW - Data augmentation

KW - Fluorescence microscopy

KW - Nuclei segmentation

UR - http://www.scopus.com/inward/record.url?scp=85023183122&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85023183122&partnerID=8YFLogxK

U2 - 10.1109/ISBI.2017.7950617

DO - 10.1109/ISBI.2017.7950617

M3 - Conference contribution

AN - SCOPUS:85023183122

SP - 704

EP - 708

BT - 2017 IEEE 14th International Symposium on Biomedical Imaging, ISBI 2017

PB - IEEE Computer Society

ER -