Deep-Learning-Based Label-Free Segmentation of Cell Nuclei in Time-Lapse Refractive Index Tomograms

Jimin Lee, Hyejin Kim, Hyungjoo Cho, Youngju Jo, Yujin Song, Daewoong Ahn, Kangwon Lee, Yongkeun Park, Sung Joon Ye

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

We proposed a method of label-free segmentation of cell nuclei by exploiting a deep learning (DL) framework. Over the years, fluorescent proteins and staining agents have been widely used to identify cell nuclei. However, the use of exogenous agents inevitably prevents from long-term imaging of live cells and rapid analysis and even interferes with intrinsic physiological conditions. Without any agents, the proposed method was applied to label-free optical diffraction tomography (ODT) of human breast cancer cells. A novel architecture with optimized training strategies was validated through cross-modality and cross-laboratory experiments. The nucleus volumes from the DL-based label-free ODT segmentation accurately agreed with those from fluorescent-based. Furthermore, the 4D cell nucleus segmentation was successfully performed for the time-lapse ODT images. The proposed method would bring out broad and immediate biomedical applications with our framework publicly available.

Original languageEnglish
Article number8743437
Pages (from-to)83449-83460
Number of pages12
JournalIEEE Access
Volume7
DOIs
StatePublished - 1 Jan 2019

Fingerprint

Labels
Refractive index
Cells
Tomography
Diffraction
Proteins
Imaging techniques
Deep learning
Experiments

Keywords

  • Cell nucleus segmentation
  • deep learning
  • label-free segmentation
  • optical diffraction tomography
  • refractive index tomogram

Cite this

Lee, Jimin ; Kim, Hyejin ; Cho, Hyungjoo ; Jo, Youngju ; Song, Yujin ; Ahn, Daewoong ; Lee, Kangwon ; Park, Yongkeun ; Ye, Sung Joon. / Deep-Learning-Based Label-Free Segmentation of Cell Nuclei in Time-Lapse Refractive Index Tomograms. In: IEEE Access. 2019 ; Vol. 7. pp. 83449-83460.
@article{8bbcf2aa18b74768a4882fcf08b97fc8,
title = "Deep-Learning-Based Label-Free Segmentation of Cell Nuclei in Time-Lapse Refractive Index Tomograms",
abstract = "We proposed a method of label-free segmentation of cell nuclei by exploiting a deep learning (DL) framework. Over the years, fluorescent proteins and staining agents have been widely used to identify cell nuclei. However, the use of exogenous agents inevitably prevents from long-term imaging of live cells and rapid analysis and even interferes with intrinsic physiological conditions. Without any agents, the proposed method was applied to label-free optical diffraction tomography (ODT) of human breast cancer cells. A novel architecture with optimized training strategies was validated through cross-modality and cross-laboratory experiments. The nucleus volumes from the DL-based label-free ODT segmentation accurately agreed with those from fluorescent-based. Furthermore, the 4D cell nucleus segmentation was successfully performed for the time-lapse ODT images. The proposed method would bring out broad and immediate biomedical applications with our framework publicly available.",
keywords = "Cell nucleus segmentation, deep learning, label-free segmentation, optical diffraction tomography, refractive index tomogram",
author = "Jimin Lee and Hyejin Kim and Hyungjoo Cho and Youngju Jo and Yujin Song and Daewoong Ahn and Kangwon Lee and Yongkeun Park and Ye, {Sung Joon}",
year = "2019",
month = "1",
day = "1",
doi = "10.1109/ACCESS.2019.2924255",
language = "English",
volume = "7",
pages = "83449--83460",
journal = "IEEE Access",
issn = "2169-3536",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

Lee, J, Kim, H, Cho, H, Jo, Y, Song, Y, Ahn, D, Lee, K, Park, Y & Ye, SJ 2019, 'Deep-Learning-Based Label-Free Segmentation of Cell Nuclei in Time-Lapse Refractive Index Tomograms', IEEE Access, vol. 7, 8743437, pp. 83449-83460. https://doi.org/10.1109/ACCESS.2019.2924255

Deep-Learning-Based Label-Free Segmentation of Cell Nuclei in Time-Lapse Refractive Index Tomograms. / Lee, Jimin; Kim, Hyejin; Cho, Hyungjoo; Jo, Youngju; Song, Yujin; Ahn, Daewoong; Lee, Kangwon; Park, Yongkeun; Ye, Sung Joon.

In: IEEE Access, Vol. 7, 8743437, 01.01.2019, p. 83449-83460.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Deep-Learning-Based Label-Free Segmentation of Cell Nuclei in Time-Lapse Refractive Index Tomograms

AU - Lee, Jimin

AU - Kim, Hyejin

AU - Cho, Hyungjoo

AU - Jo, Youngju

AU - Song, Yujin

AU - Ahn, Daewoong

AU - Lee, Kangwon

AU - Park, Yongkeun

AU - Ye, Sung Joon

PY - 2019/1/1

Y1 - 2019/1/1

N2 - We proposed a method of label-free segmentation of cell nuclei by exploiting a deep learning (DL) framework. Over the years, fluorescent proteins and staining agents have been widely used to identify cell nuclei. However, the use of exogenous agents inevitably prevents from long-term imaging of live cells and rapid analysis and even interferes with intrinsic physiological conditions. Without any agents, the proposed method was applied to label-free optical diffraction tomography (ODT) of human breast cancer cells. A novel architecture with optimized training strategies was validated through cross-modality and cross-laboratory experiments. The nucleus volumes from the DL-based label-free ODT segmentation accurately agreed with those from fluorescent-based. Furthermore, the 4D cell nucleus segmentation was successfully performed for the time-lapse ODT images. The proposed method would bring out broad and immediate biomedical applications with our framework publicly available.

AB - We proposed a method of label-free segmentation of cell nuclei by exploiting a deep learning (DL) framework. Over the years, fluorescent proteins and staining agents have been widely used to identify cell nuclei. However, the use of exogenous agents inevitably prevents from long-term imaging of live cells and rapid analysis and even interferes with intrinsic physiological conditions. Without any agents, the proposed method was applied to label-free optical diffraction tomography (ODT) of human breast cancer cells. A novel architecture with optimized training strategies was validated through cross-modality and cross-laboratory experiments. The nucleus volumes from the DL-based label-free ODT segmentation accurately agreed with those from fluorescent-based. Furthermore, the 4D cell nucleus segmentation was successfully performed for the time-lapse ODT images. The proposed method would bring out broad and immediate biomedical applications with our framework publicly available.

KW - Cell nucleus segmentation

KW - deep learning

KW - label-free segmentation

KW - optical diffraction tomography

KW - refractive index tomogram

UR - http://www.scopus.com/inward/record.url?scp=85068664425&partnerID=8YFLogxK

U2 - 10.1109/ACCESS.2019.2924255

DO - 10.1109/ACCESS.2019.2924255

M3 - Article

AN - SCOPUS:85068664425

VL - 7

SP - 83449

EP - 83460

JO - IEEE Access

JF - IEEE Access

SN - 2169-3536

M1 - 8743437

ER -