Implicit Neural Representations for Generative Modeling of
Living Cell Shapes

D 2022

Implicit Neural Representations for Generative Modeling of Living Cell Shapes

WIESNER, David, Julian SUK, Sven DUMMER, David SVOBODA, Jelmer WOLTERINK et. al.

Basic information

Original name

Implicit Neural Representations for Generative Modeling of Living Cell Shapes

Name in Czech

Implicitní neurální reprezentace pro generativní modelování tvaru živých buněk

Authors

WIESNER, David (203 Czech Republic, belonging to the institution), Julian SUK (528 Netherlands), Sven DUMMER (528 Netherlands), David SVOBODA (203 Czech Republic, guarantor, belonging to the institution) and Jelmer WOLTERINK

Edition

Switzerland, International Conference on Medical Image Computing and Computer Assisted Intervention, p. 58-67, 10 pp. 2022

Publisher

Springer Nature Switzerland

Other information

Language

English

Type of outcome

Stať ve sborníku

Field of Study

10201 Computer sciences, information science, bioinformatics

Country of publisher

Switzerland

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/00216224:14330/22:00125774

Organization unit

Faculty of Informatics

ISBN

978-3-031-16439-2

ISSN

DOI

http://dx.doi.org/10.1007/978-3-031-16440-8_6

UT WoS

000867306400006

Keywords in English

cell shape modeling; neural networks; implicit neural representations; signed distance function; generative model; interpolation

Abstract

V originále

Methods allowing the synthesis of realistic cell shapes could help generate training data sets to improve cell tracking and segmentation in biomedical images. Deep generative models for cell shape synthesis require a light-weight and flexible representation of the cell shape. However, commonly used voxel-based representations are unsuitable for high-resolution shape synthesis, and polygon meshes have limitations when modeling topology changes such as cell growth or mitosis. In this work, we propose to use level sets of signed distance functions (SDFs) to represent cell shapes. We optimize a neural network as an implicit neural representation of the SDF value at any point in a 3D+time domain. The model is conditioned on a latent code, thus allowing the synthesis of new and unseen shape sequences. We validate our approach quantitatively and qualitatively on C. elegans cells that grow and divide, and lung cancer cells with growing complex filopodial protrusions. Our results show that shape descriptors of synthetic cells resemble those of real cells, and that our model is able to generate topologically plausible sequences of complex cell shapes in 3D+time.

Links

EF18_046/0016045, research and development project

Name: Modernizace národní infrastruktury pro biologické a medicínské zobrazování Czech-BioImaging

LM2018129, research and development project

Name: Národní infrastruktura pro biologické a medicínské zobrazování Czech-BioImaging

Investor: Ministry of Education, Youth and Sports of the CR

MUNI/A/1145/2021, interní kód MU

Name: Rozsáhlé výpočetní systémy: modely, aplikace a verifikace XI. (Acronym: SV-FI MAV XI.)

Investor: Masaryk University

MUNI/A/1230/2021, interní kód MU

Name: Zapojení studentů Fakulty informatiky do mezinárodní vědecké komunity 22 (Acronym: SKOMU)

Investor: Masaryk University

MUNI/G/1446/2018, interní kód MU

Name: Deciphering the mechanisms of mammary epithelial branched pattern formation through iterative biological and mathematical modelling

Investor: Masaryk University, INTERDISCIPLINARY - Interdisciplinary research projects

Citovat

WIESNER, David, Julian SUK, Sven DUMMER, David SVOBODA and Jelmer WOLTERINK. Implicit Neural Representations for Generative Modeling of Living Cell Shapes. Online. In Linwei Wang, Qi Dou, P. Thomas Fletcher, Stefanie Speidel, Shuo Li. International Conference on Medical Image Computing and Computer Assisted Intervention. Switzerland: Springer Nature Switzerland, 2022, p. 58-67. ISBN 978-3-031-16439-2. Available from: https://dx.doi.org/10.1007/978-3-031-16440-8_6.

@inproceedings{1851540,
   author = {Wiesner, David and Suk, Julian and Dummer, Sven and Svoboda, David and Wolterink, Jelmer},
   address = {Switzerland},
   booktitle = {International Conference on Medical Image Computing and Computer Assisted Intervention},
   doi = {http://dx.doi.org/10.1007/978-3-031-16440-8_6},
   editor = {Linwei Wang, Qi Dou, P. Thomas Fletcher, Stefanie Speidel, Shuo Li},
   keywords = {cell shape modeling; neural networks; implicit neural representations; signed distance function; generative model; interpolation},
   howpublished = {elektronická verze "online"},
   language = {eng},
   location = {Switzerland},
   isbn = {978-3-031-16439-2},
   pages = {58-67},
   publisher = {Springer Nature Switzerland},
   title = {Implicit Neural Representations for Generative Modeling of Living Cell Shapes},
   url = {http://dx.doi.org/10.1007/978-3-031-16440-8_6},
   year = {2022}
}

TY  - JOUR
ID  - 1851540
AU  - Wiesner, David - Suk, Julian - Dummer, Sven - Svoboda, David - Wolterink, Jelmer
PY  - 2022
TI  - Implicit Neural Representations for Generative Modeling of Living Cell Shapes
PB  - Springer Nature Switzerland
CY  - Switzerland
SN  - 9783031164392
KW  - cell shape modeling
KW  - neural networks
KW  - implicit neural representations
KW  - signed distance function
KW  - generative model
KW  - interpolation
UR  - http://dx.doi.org/10.1007/978-3-031-16440-8_6
N2  - Methods allowing the synthesis of realistic cell shapes could help generate training data sets to improve cell tracking and segmentation in biomedical images. Deep generative models for cell shape synthesis require a light-weight and flexible representation of the cell shape. However, commonly used voxel-based representations are unsuitable for high-resolution shape synthesis, and polygon meshes have limitations when modeling topology changes such as cell growth or mitosis. In this work, we propose to use level sets of signed distance functions (SDFs) to represent cell shapes. We optimize a neural network as an implicit neural representation of the SDF value at any point in a 3D+time domain. The model is conditioned on a latent code, thus allowing the synthesis of new and unseen shape sequences. We validate our approach quantitatively and qualitatively on C. elegans cells that grow and divide, and lung cancer cells with growing complex filopodial protrusions. Our results show that shape descriptors of synthetic cells resemble those of real cells, and that our model is able to generate topologically plausible sequences of complex cell shapes in 3D+time.
ER  -

WIESNER, David, Julian SUK, Sven DUMMER, David SVOBODA and Jelmer WOLTERINK. Implicit Neural Representations for Generative Modeling of Living Cell Shapes. Online. In Linwei Wang, Qi Dou, P. Thomas Fletcher, Stefanie Speidel, Shuo Li. \textit{International Conference on Medical Image Computing and Computer Assisted Intervention}. Switzerland: Springer Nature Switzerland, 2022, p.~58-67. ISBN~978-3-031-16439-2. Available from: https://dx.doi.org/10.1007/978-3-031-16440-8\_{}6.

Detailed Information on Publication Record