TY - JOUR
T1 - Fast and robust multiplane single-molecule localization microscopy using a deep neural network
AU - Aritake, Toshimitsu
AU - Hino, Hideitsu
AU - Namiki, Shigeyuki
AU - Asanuma, Daisuke
AU - Hirose, Kenzo
AU - Murata, Noboru
N1 - Funding Information:
The authors would like to thank M. Tanaka for technical assistance. This work was partially supported by JSPS KAKENHI (Grant Nos. 17H01793 and 18H03291) and JST CREST (Grant Nos. JPMJCR1761 and JPMJCR14D7).
Funding Information:
Dr. Noboru Murata reports grants from Japan Society for the Promotion of Science KAKENHI Grant Numbers 17H01793 , 18H03291 and grants from Japan Science and Technology Agency CREST Grant.
Funding Information:
The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Dr. Hideitsu Hino reports grants from Japan Society for the Promotion of Science KAKENHI Grant Numbers 17H01793 and grants from Japan Science and Technology Agency CREST Grant Number JPMJCR1761 during the conduct of the study.
Publisher Copyright:
© 2021 The Authors
PY - 2021/9/3
Y1 - 2021/9/3
N2 - Single-molecule localization microscopy is a widely used technique in biological research for measuring the nanostructures of samples smaller than the diffraction limit. This study uses multifocal plane microscopy and addresses the three-dimensional (3D) single-molecule localization problem, where lateral and axial locations of molecules are estimated. However, when multifocal plane microscopy is used, the estimation accuracy of 3D localization is easily deteriorated by the small lateral drifts of camera positions. A 3D molecule localization problem was presented along with the lateral drift estimation as a compressed sensing problem. A deep neural network (DNN) was applied to solve this problem accurately and efficiently. The results show that the proposed method is robust to lateral drift and achieves an accuracy of 20 nm laterally and 50 nm axially without an explicit drift correction.
AB - Single-molecule localization microscopy is a widely used technique in biological research for measuring the nanostructures of samples smaller than the diffraction limit. This study uses multifocal plane microscopy and addresses the three-dimensional (3D) single-molecule localization problem, where lateral and axial locations of molecules are estimated. However, when multifocal plane microscopy is used, the estimation accuracy of 3D localization is easily deteriorated by the small lateral drifts of camera positions. A 3D molecule localization problem was presented along with the lateral drift estimation as a compressed sensing problem. A deep neural network (DNN) was applied to solve this problem accurately and efficiently. The results show that the proposed method is robust to lateral drift and achieves an accuracy of 20 nm laterally and 50 nm axially without an explicit drift correction.
KW - 3D single-molecule localization microscopy
KW - Convolutional neural network
KW - Lateral drift
KW - Multi-focal plane microscopy
UR - http://www.scopus.com/inward/record.url?scp=85105511848&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85105511848&partnerID=8YFLogxK
U2 - 10.1016/j.neucom.2021.04.050
DO - 10.1016/j.neucom.2021.04.050
M3 - Article
AN - SCOPUS:85105511848
SN - 0925-2312
VL - 451
SP - 279
EP - 289
JO - Neurocomputing
JF - Neurocomputing
ER -