On Generative Modeling of Cell Shape Using 3D GANs

D 2019

On Generative Modeling of Cell Shape Using 3D GANs

WIESNER, David, Tereza NEČASOVÁ and David SVOBODA

Basic information

Original name

On Generative Modeling of Cell Shape Using 3D GANs

Authors

WIESNER, David (203 Czech Republic, guarantor, belonging to the institution), Tereza NEČASOVÁ (203 Czech Republic, belonging to the institution) and David SVOBODA (203 Czech Republic, belonging to the institution)

Edition

LNCS 11752. Trento, Image Analysis and Processing – ICIAP 2019, p. 672-682, 11 pp. 2019

Publisher

Springer

Other information

Language

English

Type of outcome

Stať ve sborníku

Field of Study

10201 Computer sciences, information science, bioinformatics

Country of publisher

United States of America

Confidentiality degree

není předmětem státního či obchodního tajemství

Publication form

electronic version available online

References:

URL

Impact factor

Impact factor: 0.402 in 2005

RIV identification code

RIV/00216224:14330/19:00107522

Organization unit

Faculty of Informatics

ISBN

978-3-030-30644-1

ISSN

DOI

http://dx.doi.org/10.1007/978-3-030-30645-8_61

UT WoS

000562008400059

Keywords in English

Image-based Simulations; 3D GAN; Training Stability; Microscopy Data; Digital Cell Shape

Abstract

V originále

The ongoing advancement of deep-learning generative models, showing great interest of the scientific community since the introduction of the generative adversarial networks (GAN), paved the way for generation of realistic data. The utilization of deep learning for the generation of realistic biomedical images allows one to alleviate the constraints of the parametric models, limited by the employed mathematical approximations. Building further upon the laid foundation, the 3D GAN added another dimension, allowing generation of fully 3D volumetric data. In this paper, we present an approach to generating fully 3D volumetric cell masks using GANs. Presented model is able to generate high-quality cell masks with variability matching the real data. Required modifications of the proposed model are presented along with the training dataset, based on 385 real cells captured using the fluorescence microscope. Furthermore, the statistical validation is also presented, allowing to quantitatively assess the quality of data generated by the proposed model.

Links

GA17-05048S, research and development project

Name: Segmentace a trekování živých buněk v multimodálních obrazech

Investor: Czech Science Foundation

MUNI/A/1018/2018, interní kód MU

Name: Rozsáhlé výpočetní systémy: modely, aplikace a verifikace VIII.

Investor: Masaryk University, Category A

Citovat

WIESNER, David, Tereza NEČASOVÁ and David SVOBODA. On Generative Modeling of Cell Shape Using 3D GANs. Online. In Ricci Elisa, Rota Buló Samuel, Snoek Cees, Lanz Oswald, Messelodi Stefano, Sebe Nicu. Image Analysis and Processing – ICIAP 2019. LNCS 11752. Trento: Springer, 2019, p. 672-682. ISBN 978-3-030-30644-1. Available from: https://dx.doi.org/10.1007/978-3-030-30645-8_61.

@inproceedings{1546941,
   author = {Wiesner, David and Nečasová, Tereza and Svoboda, David},
   address = {Trento},
   booktitle = {Image Analysis and Processing – ICIAP 2019},
   doi = {http://dx.doi.org/10.1007/978-3-030-30645-8_61},
   edition = {LNCS 11752},
   editor = {Ricci Elisa, Rota Buló Samuel, Snoek Cees, Lanz Oswald, Messelodi Stefano, Sebe Nicu},
   keywords = {Image-based Simulations; 3D GAN; Training Stability; Microscopy Data; Digital Cell Shape},
   howpublished = {elektronická verze "online"},
   language = {eng},
   location = {Trento},
   isbn = {978-3-030-30644-1},
   pages = {672-682},
   publisher = {Springer},
   title = {On Generative Modeling of Cell Shape Using 3D GANs},
   url = {http://dx.doi.org/10.1007/978-3-030-30645-8_61},
   year = {2019}
}

TY  - JOUR
ID  - 1546941
AU  - Wiesner, David - Nečasová, Tereza - Svoboda, David
PY  - 2019
TI  - On Generative Modeling of Cell Shape Using 3D GANs
PB  - Springer
CY  - Trento
SN  - 9783030306441
KW  - Image-based Simulations
KW  - 3D GAN
KW  - Training Stability
KW  - Microscopy Data
KW  - Digital Cell Shape
UR  - http://dx.doi.org/10.1007/978-3-030-30645-8_61
N2  - The ongoing advancement of deep-learning generative models, showing great interest of the scientific community since the introduction of the generative adversarial networks (GAN), paved the way for generation of realistic data. The utilization of deep learning for the generation of realistic biomedical images allows one to alleviate the constraints of the parametric models, limited by the employed mathematical approximations. Building further upon the laid foundation, the 3D GAN added another dimension, allowing generation of fully 3D volumetric data. In this paper, we present an approach to generating fully 3D volumetric cell masks using GANs. Presented model is able to generate high-quality cell masks with variability matching the real data. Required modifications of the proposed model are presented along with the training dataset, based on 385 real cells captured using the fluorescence microscope. Furthermore, the statistical validation is also presented, allowing to quantitatively assess the quality of data generated by the proposed model.
ER  -

WIESNER, David, Tereza NEČASOVÁ and David SVOBODA. On Generative Modeling of Cell Shape Using 3D GANs. Online. In Ricci Elisa, Rota Buló Samuel, Snoek Cees, Lanz Oswald, Messelodi Stefano, Sebe Nicu. \textit{Image Analysis and Processing – ICIAP 2019}. LNCS 11752. Trento: Springer, 2019, p.~672-682. ISBN~978-3-030-30644-1. Available from: https://dx.doi.org/10.1007/978-3-030-30645-8\_{}61.

Detailed Information on Publication Record