Field-dependent deep learning enables high-throughput whole-cell 3D super-resolution imaging.

Fu, S; Shi, W; Luo, T; He, Y; Zhou, L; Yang, J; Yang, Z; Liu, J; Liu, X; Guo, Z; Yang, C; Liu, C; Huang, Z-L; Ries, J; Zhang, M; Xi, P; Jin, D; Li, Y

Field-dependent deep learning enables high-throughput whole-cell 3D super-resolution imaging.

Fu, S Shi, W Luo, T He, Y Zhou, L Yang, J Yang, Z Liu, J Liu, X Guo, Z Yang, C Liu, C Huang, Z-L Ries, J Zhang, M Xi, P Jin, D

Li, Y

Permalink

Publisher:: NATURE PORTFOLIO
Publication Type:: Journal Article
Citation:: Nat Methods, 2023, 20, (3), pp. 459-468
Issue Date:: 2023-03

Closed Access

	Filename	Description	Size
	s41592-023-01775-5.pdf	Published version	31.86 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Fu, S
dc.contributor.author	Shi, W
dc.contributor.author	Luo, T
dc.contributor.author	He, Y
dc.contributor.author	Zhou, L
dc.contributor.author	Yang, J
dc.contributor.author	Yang, Z
dc.contributor.author	Liu, J
dc.contributor.author	Liu, X
dc.contributor.author	Guo, Z
dc.contributor.author	Yang, C
dc.contributor.author	Liu, C
dc.contributor.author	Huang, Z-L
dc.contributor.author	Ries, J
dc.contributor.author	Zhang, M
dc.contributor.author	Xi, P
dc.contributor.author	Jin, D https://orcid.org/0000-0003-1046-2666
dc.contributor.author	Li, Y
dc.date.accessioned	2024-02-09T02:42:16Z
dc.date.available	2023-01-09
dc.date.available	2024-02-09T02:42:16Z
dc.date.issued	2023-03
dc.identifier.citation	Nat Methods, 2023, 20, (3), pp. 459-468
dc.identifier.issn	1548-7091
dc.identifier.issn	1548-7105
dc.identifier.uri	http://hdl.handle.net/10453/175540
dc.description.abstract	Single-molecule localization microscopy in a typical wide-field setup has been widely used for investigating subcellular structures with super resolution; however, field-dependent aberrations restrict the field of view (FOV) to only tens of micrometers. Here, we present a deep-learning method for precise localization of spatially variant point emitters (FD-DeepLoc) over a large FOV covering the full chip of a modern sCMOS camera. Using a graphic processing unit-based vectorial point spread function (PSF) fitter, we can fast and accurately model the spatially variant PSF of a high numerical aperture objective in the entire FOV. Combined with deformable mirror-based optimal PSF engineering, we demonstrate high-accuracy three-dimensional single-molecule localization microscopy over a volume of ~180 × 180 × 5 μm3, allowing us to image mitochondria and nuclear pore complexes in entire cells in a single imaging cycle without hardware scanning; a 100-fold increase in throughput compared to the state of the art.
dc.format	Print-Electronic
dc.language	eng
dc.publisher	NATURE PORTFOLIO
dc.relation.ispartof	Nat Methods
dc.relation.isbasedon	10.1038/s41592-023-01775-5
dc.rights	info:eu-repo/semantics/closedAccess
dc.subject	06 Biological Sciences, 10 Technology, 11 Medical and Health Sciences
dc.subject.classification	Developmental Biology
dc.subject.classification	31 Biological sciences
dc.subject.mesh	Deep Learning
dc.subject.mesh	Imaging, Three-Dimensional
dc.subject.mesh	Single Molecule Imaging
dc.subject.mesh	Imaging, Three-Dimensional
dc.subject.mesh	Single Molecule Imaging
dc.subject.mesh	Deep Learning
dc.subject.mesh	Deep Learning
dc.subject.mesh	Imaging, Three-Dimensional
dc.subject.mesh	Single Molecule Imaging
dc.title	Field-dependent deep learning enables high-throughput whole-cell 3D super-resolution imaging.
dc.type	Journal Article
utslib.citation.volume	20
utslib.location.activity	United States
utslib.for	06 Biological Sciences
utslib.for	10 Technology
utslib.for	11 Medical and Health Sciences
pubs.organisational-group	University of Technology Sydney
pubs.organisational-group	University of Technology Sydney/Faculty of Science
pubs.organisational-group	University of Technology Sydney/Faculty of Science/School of Mathematical and Physical Sciences
pubs.organisational-group	University of Technology Sydney/Strength - IBMD - Initiative for Biomedical Devices
utslib.copyright.status	closed_access	*
dc.date.updated	2024-02-09T02:42:06Z
pubs.issue	3
pubs.publication-status	Published
pubs.volume	20
utslib.citation.issue	3

Abstract:

Single-molecule localization microscopy in a typical wide-field setup has been widely used for investigating subcellular structures with super resolution; however, field-dependent aberrations restrict the field of view (FOV) to only tens of micrometers. Here, we present a deep-learning method for precise localization of spatially variant point emitters (FD-DeepLoc) over a large FOV covering the full chip of a modern sCMOS camera. Using a graphic processing unit-based vectorial point spread function (PSF) fitter, we can fast and accurately model the spatially variant PSF of a high numerical aperture objective in the entire FOV. Combined with deformable mirror-based optimal PSF engineering, we demonstrate high-accuracy three-dimensional single-molecule localization microscopy over a volume of ~180 × 180 × 5 μm3, allowing us to image mitochondria and nuclear pore complexes in entire cells in a single imaging cycle without hardware scanning; a 100-fold increase in throughput compared to the state of the art.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/175540