Variational autoencoding tissue response to microenvironment perturbation

Geoffrey F. Schau; Guillaume Thibault; Mark A. Dane; Joe W. Gray; Laura M. Heiser; Young Hwan Chang

doi:10.1117/12.2512660

Variational autoencoding tissue response to microenvironment perturbation

Geoffrey F. Schau, Guillaume Thibault, Mark A. Dane, Joe W. Gray, Laura M. Heiser, Young Hwan Chang

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

1 Scopus citations

Abstract

This work applies deep variational autoencoder learning architecture to study multi-cellular growth characteristics of human mammary epithelial cells in response to diverse microenvironment perturbations. Our approach introduces a novel method of visualizing learned feature spaces of trained variational autoencoding models that enables visualization of principal features in two dimensions. We find that unsupervised learned features more closely associate with expert annotation of cell colony organization than biologically-inspired hand-crafted features, demonstrating the utility of deep learning systems to meaningfully characterize features of multi-cellular growth characteristics in a fully unsupervised and data-driven manner.

Original language	English (US)
Title of host publication	Medical Imaging 2019
Subtitle of host publication	Image Processing
Editors	Elsa D. Angelini, Elsa D. Angelini, Elsa D. Angelini, Bennett A. Landman
Publisher	SPIE
ISBN (Electronic)	9781510625457
DOIs	https://doi.org/10.1117/12.2512660
State	Published - 2019
Event	Medical Imaging 2019: Image Processing - San Diego, United States Duration: Feb 19 2019 → Feb 21 2019

Publication series

Name	Progress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume	10949
ISSN (Print)	1605-7422

Conference

Conference	Medical Imaging 2019: Image Processing
Country/Territory	United States
City	San Diego
Period	2/19/19 → 2/21/19

ASJC Scopus subject areas

Electronic, Optical and Magnetic Materials
Atomic and Molecular Physics, and Optics
Radiology Nuclear Medicine and imaging
Biomaterials

Access to Document

10.1117/12.2512660

Cite this

Schau, G. F., Thibault, G., Dane, M. A., Gray, J. W., Heiser, L. M., & Chang, Y. H. (2019). Variational autoencoding tissue response to microenvironment perturbation. In E. D. Angelini, E. D. Angelini, E. D. Angelini, & B. A. Landman (Eds.), Medical Imaging 2019: Image Processing Article 109491M (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 10949). SPIE. https://doi.org/10.1117/12.2512660

Variational autoencoding tissue response to microenvironment perturbation. / Schau, Geoffrey F.; Thibault, Guillaume; Dane, Mark A. et al.
Medical Imaging 2019: Image Processing. ed. / Elsa D. Angelini; Elsa D. Angelini; Elsa D. Angelini; Bennett A. Landman. SPIE, 2019. 109491M (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 10949).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Schau, GF, Thibault, G, Dane, MA, Gray, JW, Heiser, LM & Chang, YH 2019, Variational autoencoding tissue response to microenvironment perturbation. in ED Angelini, ED Angelini, ED Angelini & BA Landman (eds), Medical Imaging 2019: Image Processing., 109491M, Progress in Biomedical Optics and Imaging - Proceedings of SPIE, vol. 10949, SPIE, Medical Imaging 2019: Image Processing, San Diego, United States, 2/19/19. https://doi.org/10.1117/12.2512660

@inproceedings{7cb1088538c04aafbb5ca06133e65126,

title = "Variational autoencoding tissue response to microenvironment perturbation",

abstract = "This work applies deep variational autoencoder learning architecture to study multi-cellular growth characteristics of human mammary epithelial cells in response to diverse microenvironment perturbations. Our approach introduces a novel method of visualizing learned feature spaces of trained variational autoencoding models that enables visualization of principal features in two dimensions. We find that unsupervised learned features more closely associate with expert annotation of cell colony organization than biologically-inspired hand-crafted features, demonstrating the utility of deep learning systems to meaningfully characterize features of multi-cellular growth characteristics in a fully unsupervised and data-driven manner.",

author = "Schau, {Geoffrey F.} and Guillaume Thibault and Dane, {Mark A.} and Gray, {Joe W.} and Heiser, {Laura M.} and Chang, {Young Hwan}",

note = "Publisher Copyright: {\textcopyright} COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only.; Medical Imaging 2019: Image Processing ; Conference date: 19-02-2019 Through 21-02-2019",

year = "2019",

doi = "10.1117/12.2512660",

language = "English (US)",

series = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",

publisher = "SPIE",

editor = "Angelini, {Elsa D.} and Angelini, {Elsa D.} and Angelini, {Elsa D.} and Landman, {Bennett A.}",

booktitle = "Medical Imaging 2019",

}

TY - GEN

T1 - Variational autoencoding tissue response to microenvironment perturbation

AU - Schau, Geoffrey F.

AU - Thibault, Guillaume

AU - Dane, Mark A.

AU - Gray, Joe W.

AU - Heiser, Laura M.

AU - Chang, Young Hwan

PY - 2019

Y1 - 2019

N2 - This work applies deep variational autoencoder learning architecture to study multi-cellular growth characteristics of human mammary epithelial cells in response to diverse microenvironment perturbations. Our approach introduces a novel method of visualizing learned feature spaces of trained variational autoencoding models that enables visualization of principal features in two dimensions. We find that unsupervised learned features more closely associate with expert annotation of cell colony organization than biologically-inspired hand-crafted features, demonstrating the utility of deep learning systems to meaningfully characterize features of multi-cellular growth characteristics in a fully unsupervised and data-driven manner.

AB - This work applies deep variational autoencoder learning architecture to study multi-cellular growth characteristics of human mammary epithelial cells in response to diverse microenvironment perturbations. Our approach introduces a novel method of visualizing learned feature spaces of trained variational autoencoding models that enables visualization of principal features in two dimensions. We find that unsupervised learned features more closely associate with expert annotation of cell colony organization than biologically-inspired hand-crafted features, demonstrating the utility of deep learning systems to meaningfully characterize features of multi-cellular growth characteristics in a fully unsupervised and data-driven manner.

UR - http://www.scopus.com/inward/record.url?scp=85068319226&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85068319226&partnerID=8YFLogxK

U2 - 10.1117/12.2512660

DO - 10.1117/12.2512660

M3 - Conference contribution

AN - SCOPUS:85068319226

T3 - Progress in Biomedical Optics and Imaging - Proceedings of SPIE

BT - Medical Imaging 2019

A2 - Angelini, Elsa D.

A2 - Landman, Bennett A.

PB - SPIE

T2 - Medical Imaging 2019: Image Processing

Y2 - 19 February 2019 through 21 February 2019

ER -

Variational autoencoding tissue response to microenvironment perturbation

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this