Predicting the future direction of cell movement with convolutional neural networks

Shori Nishimoto, Yuta Tokuoka, Takahiro Yamada, Noriko F. Hiroi, Akira Funahashi

Research output: Contribution to journalArticle

Abstract

Image-based deep learning systems, such as convolutional neural networks (CNNs), have recently been applied to cell classification, producing impressive results; however, application of CNNs has been confined to classification of the current cell state from the image. Here, we focused on cell movement where current and/or past cell shape can influence the future cell movement. We demonstrate that CNNs prospectively predicted the future direction of cell movement with high accuracy from a single image patch of a cell at a certain time. Furthermore, by visualizing the image features that were learned by the CNNs, we could identify morphological features, e.g., the protrusions and trailing edge that have been experimentally reported to determine the direction of cell movement. Our results indicate that CNNs have the potential to predict the future direction of cell movement from current cell shape, and can be used to automatically identify those morphological features that influence future cell movement.

Original languageEnglish
Article numbere0221245
JournalPloS one
Volume14
Issue number9
DOIs
Publication statusPublished - 2019 Jan 1

Fingerprint

cell movement
neural networks
Cell Movement
Cells
Neural networks
Cell Shape
cells
Learning systems
Direction compound
learning
Learning

ASJC Scopus subject areas

  • Biochemistry, Genetics and Molecular Biology(all)
  • Agricultural and Biological Sciences(all)

Cite this

Predicting the future direction of cell movement with convolutional neural networks. / Nishimoto, Shori; Tokuoka, Yuta; Yamada, Takahiro; Hiroi, Noriko F.; Funahashi, Akira.

In: PloS one, Vol. 14, No. 9, e0221245, 01.01.2019.

Research output: Contribution to journalArticle

@article{4ca46c772bfb4ca7ae231b9840938616,
title = "Predicting the future direction of cell movement with convolutional neural networks",
abstract = "Image-based deep learning systems, such as convolutional neural networks (CNNs), have recently been applied to cell classification, producing impressive results; however, application of CNNs has been confined to classification of the current cell state from the image. Here, we focused on cell movement where current and/or past cell shape can influence the future cell movement. We demonstrate that CNNs prospectively predicted the future direction of cell movement with high accuracy from a single image patch of a cell at a certain time. Furthermore, by visualizing the image features that were learned by the CNNs, we could identify morphological features, e.g., the protrusions and trailing edge that have been experimentally reported to determine the direction of cell movement. Our results indicate that CNNs have the potential to predict the future direction of cell movement from current cell shape, and can be used to automatically identify those morphological features that influence future cell movement.",
author = "Shori Nishimoto and Yuta Tokuoka and Takahiro Yamada and Hiroi, {Noriko F.} and Akira Funahashi",
year = "2019",
month = "1",
day = "1",
doi = "10.1371/journal.pone.0221245",
language = "English",
volume = "14",
journal = "PLoS One",
issn = "1932-6203",
publisher = "Public Library of Science",
number = "9",

}

TY - JOUR

T1 - Predicting the future direction of cell movement with convolutional neural networks

AU - Nishimoto, Shori

AU - Tokuoka, Yuta

AU - Yamada, Takahiro

AU - Hiroi, Noriko F.

AU - Funahashi, Akira

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Image-based deep learning systems, such as convolutional neural networks (CNNs), have recently been applied to cell classification, producing impressive results; however, application of CNNs has been confined to classification of the current cell state from the image. Here, we focused on cell movement where current and/or past cell shape can influence the future cell movement. We demonstrate that CNNs prospectively predicted the future direction of cell movement with high accuracy from a single image patch of a cell at a certain time. Furthermore, by visualizing the image features that were learned by the CNNs, we could identify morphological features, e.g., the protrusions and trailing edge that have been experimentally reported to determine the direction of cell movement. Our results indicate that CNNs have the potential to predict the future direction of cell movement from current cell shape, and can be used to automatically identify those morphological features that influence future cell movement.

AB - Image-based deep learning systems, such as convolutional neural networks (CNNs), have recently been applied to cell classification, producing impressive results; however, application of CNNs has been confined to classification of the current cell state from the image. Here, we focused on cell movement where current and/or past cell shape can influence the future cell movement. We demonstrate that CNNs prospectively predicted the future direction of cell movement with high accuracy from a single image patch of a cell at a certain time. Furthermore, by visualizing the image features that were learned by the CNNs, we could identify morphological features, e.g., the protrusions and trailing edge that have been experimentally reported to determine the direction of cell movement. Our results indicate that CNNs have the potential to predict the future direction of cell movement from current cell shape, and can be used to automatically identify those morphological features that influence future cell movement.

UR - http://www.scopus.com/inward/record.url?scp=85071770845&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071770845&partnerID=8YFLogxK

U2 - 10.1371/journal.pone.0221245

DO - 10.1371/journal.pone.0221245

M3 - Article

C2 - 31483827

AN - SCOPUS:85071770845

VL - 14

JO - PLoS One

JF - PLoS One

SN - 1932-6203

IS - 9

M1 - e0221245

ER -