Journal article 1191 views 187 downloads
Reconstructing cell cycle and disease progression using deep learning
Philipp Eulenberg,
Niklas Köhler,
Thomas Blasi,
Andrew Filby,
Paul Rees ,
Anne E. Carpenter,
Fabian J. Theis,
F. Alexander Wolf
Nature Communications, Volume: 8, Issue: 1
Swansea University Author: Paul Rees
-
PDF | Version of Record
Distributed under the terms of a Creative Commons Attribution (CC-BY) Licence.
Download (1.38MB)
DOI (Published version): 10.1038/s41467-017-00623-3
Abstract
We show that deep convolutional neural networks combined with nonlinear dimension reduction enable reconstructing biological processes based on raw image data. We demonstrate this by reconstructing the cell cycle of Jurkat cells and disease progression in diabetic retinopathy. In further analysis of...
Published in: | Nature Communications |
---|---|
ISSN: | 2041-1723 2041-1723 |
Published: |
2017
|
Online Access: |
Check full text
|
URI: | https://cronfa.swan.ac.uk/Record/cronfa35476 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
first_indexed |
2017-09-21T19:04:55Z |
---|---|
last_indexed |
2021-01-15T03:55:28Z |
id |
cronfa35476 |
recordtype |
SURis |
fullrecord |
<?xml version="1.0"?><rfc1807><datestamp>2021-01-14T13:14:28.3901487</datestamp><bib-version>v2</bib-version><id>35476</id><entry>2017-09-21</entry><title>Reconstructing cell cycle and disease progression using deep learning</title><swanseaauthors><author><sid>537a2fe031a796a3bde99679ee8c24f5</sid><ORCID>0000-0002-7715-6914</ORCID><firstname>Paul</firstname><surname>Rees</surname><name>Paul Rees</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2017-09-21</date><deptcode>MEDE</deptcode><abstract>We show that deep convolutional neural networks combined with nonlinear dimension reduction enable reconstructing biological processes based on raw image data. We demonstrate this by reconstructing the cell cycle of Jurkat cells and disease progression in diabetic retinopathy. In further analysis of Jurkat cells, we detect and separate a subpopulation of dead cells in an unsupervised manner and, in classifying discrete cell cycle stages, we reach a sixfold reduction in error rate compared to a recent approach based on boosting on image features. In contrast to previous methods, deep learning based predictions are fast enough for on-the-fly analysis in an imaging flow cytometer.</abstract><type>Journal Article</type><journal>Nature Communications</journal><volume>8</volume><journalNumber>1</journalNumber><paginationStart/><paginationEnd/><publisher/><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint>2041-1723</issnPrint><issnElectronic>2041-1723</issnElectronic><keywords>Cell division, Image processing, Machine learning</keywords><publishedDay>1</publishedDay><publishedMonth>12</publishedMonth><publishedYear>2017</publishedYear><publishedDate>2017-12-01</publishedDate><doi>10.1038/s41467-017-00623-3</doi><url/><notes>2018: This research is the first demonstration of deep learning applied to high throughput single cell image data, we demonstrate we can ‘learn’ cell cycle without using fluorescent cell markers and showed deep learning can define disease progression in an unsupervised manner. The work was undertaken as part of a BBSRC (BB/N005163/1) grant and also a USA NSF grant (Award Number 1458626 - Paul Rees co-investigator as visiting Professor at the Broad Institute of MIT and Harvard). It further led to another BBSRC (BB/P026818/1) grant which aims to make these tools open access.</notes><college>COLLEGE NANME</college><department>Biomedical Engineering</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>MEDE</DepartmentCode><institution>Swansea University</institution><apcterm/><lastEdited>2021-01-14T13:14:28.3901487</lastEdited><Created>2017-09-21T14:37:52.7622143</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Engineering and Applied Sciences - Biomedical Engineering</level></path><authors><author><firstname>Philipp</firstname><surname>Eulenberg</surname><order>1</order></author><author><firstname>Niklas</firstname><surname>Köhler</surname><order>2</order></author><author><firstname>Thomas</firstname><surname>Blasi</surname><order>3</order></author><author><firstname>Andrew</firstname><surname>Filby</surname><order>4</order></author><author><firstname>Paul</firstname><surname>Rees</surname><orcid>0000-0002-7715-6914</orcid><order>5</order></author><author><firstname>Anne E.</firstname><surname>Carpenter</surname><order>6</order></author><author><firstname>Fabian J.</firstname><surname>Theis</surname><order>7</order></author><author><firstname>F. Alexander</firstname><surname>Wolf</surname><order>8</order></author></authors><documents><document><filename>0035476-21092017144020.pdf</filename><originalFilename>eulenberg2017.pdf</originalFilename><uploaded>2017-09-21T14:40:20.2170000</uploaded><type>Output</type><contentLength>1493362</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><embargoDate>2017-09-21T00:00:00.0000000</embargoDate><documentNotes>Distributed under the terms of a Creative Commons Attribution (CC-BY) Licence.</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>https://creativecommons.org/licenses/by/4.0/</licence></document></documents><OutputDurs/></rfc1807> |
spelling |
2021-01-14T13:14:28.3901487 v2 35476 2017-09-21 Reconstructing cell cycle and disease progression using deep learning 537a2fe031a796a3bde99679ee8c24f5 0000-0002-7715-6914 Paul Rees Paul Rees true false 2017-09-21 MEDE We show that deep convolutional neural networks combined with nonlinear dimension reduction enable reconstructing biological processes based on raw image data. We demonstrate this by reconstructing the cell cycle of Jurkat cells and disease progression in diabetic retinopathy. In further analysis of Jurkat cells, we detect and separate a subpopulation of dead cells in an unsupervised manner and, in classifying discrete cell cycle stages, we reach a sixfold reduction in error rate compared to a recent approach based on boosting on image features. In contrast to previous methods, deep learning based predictions are fast enough for on-the-fly analysis in an imaging flow cytometer. Journal Article Nature Communications 8 1 2041-1723 2041-1723 Cell division, Image processing, Machine learning 1 12 2017 2017-12-01 10.1038/s41467-017-00623-3 2018: This research is the first demonstration of deep learning applied to high throughput single cell image data, we demonstrate we can ‘learn’ cell cycle without using fluorescent cell markers and showed deep learning can define disease progression in an unsupervised manner. The work was undertaken as part of a BBSRC (BB/N005163/1) grant and also a USA NSF grant (Award Number 1458626 - Paul Rees co-investigator as visiting Professor at the Broad Institute of MIT and Harvard). It further led to another BBSRC (BB/P026818/1) grant which aims to make these tools open access. COLLEGE NANME Biomedical Engineering COLLEGE CODE MEDE Swansea University 2021-01-14T13:14:28.3901487 2017-09-21T14:37:52.7622143 Faculty of Science and Engineering School of Engineering and Applied Sciences - Biomedical Engineering Philipp Eulenberg 1 Niklas Köhler 2 Thomas Blasi 3 Andrew Filby 4 Paul Rees 0000-0002-7715-6914 5 Anne E. Carpenter 6 Fabian J. Theis 7 F. Alexander Wolf 8 0035476-21092017144020.pdf eulenberg2017.pdf 2017-09-21T14:40:20.2170000 Output 1493362 application/pdf Version of Record true 2017-09-21T00:00:00.0000000 Distributed under the terms of a Creative Commons Attribution (CC-BY) Licence. true eng https://creativecommons.org/licenses/by/4.0/ |
title |
Reconstructing cell cycle and disease progression using deep learning |
spellingShingle |
Reconstructing cell cycle and disease progression using deep learning Paul Rees |
title_short |
Reconstructing cell cycle and disease progression using deep learning |
title_full |
Reconstructing cell cycle and disease progression using deep learning |
title_fullStr |
Reconstructing cell cycle and disease progression using deep learning |
title_full_unstemmed |
Reconstructing cell cycle and disease progression using deep learning |
title_sort |
Reconstructing cell cycle and disease progression using deep learning |
author_id_str_mv |
537a2fe031a796a3bde99679ee8c24f5 |
author_id_fullname_str_mv |
537a2fe031a796a3bde99679ee8c24f5_***_Paul Rees |
author |
Paul Rees |
author2 |
Philipp Eulenberg Niklas Köhler Thomas Blasi Andrew Filby Paul Rees Anne E. Carpenter Fabian J. Theis F. Alexander Wolf |
format |
Journal article |
container_title |
Nature Communications |
container_volume |
8 |
container_issue |
1 |
publishDate |
2017 |
institution |
Swansea University |
issn |
2041-1723 2041-1723 |
doi_str_mv |
10.1038/s41467-017-00623-3 |
college_str |
Faculty of Science and Engineering |
hierarchytype |
|
hierarchy_top_id |
facultyofscienceandengineering |
hierarchy_top_title |
Faculty of Science and Engineering |
hierarchy_parent_id |
facultyofscienceandengineering |
hierarchy_parent_title |
Faculty of Science and Engineering |
department_str |
School of Engineering and Applied Sciences - Biomedical Engineering{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Engineering and Applied Sciences - Biomedical Engineering |
document_store_str |
1 |
active_str |
0 |
description |
We show that deep convolutional neural networks combined with nonlinear dimension reduction enable reconstructing biological processes based on raw image data. We demonstrate this by reconstructing the cell cycle of Jurkat cells and disease progression in diabetic retinopathy. In further analysis of Jurkat cells, we detect and separate a subpopulation of dead cells in an unsupervised manner and, in classifying discrete cell cycle stages, we reach a sixfold reduction in error rate compared to a recent approach based on boosting on image features. In contrast to previous methods, deep learning based predictions are fast enough for on-the-fly analysis in an imaging flow cytometer. |
published_date |
2017-12-01T03:44:09Z |
_version_ |
1763752069382012928 |
score |
11.024221 |