Home > Research > Publications & Outputs > Deep learning-based aberration compensation imp...

Research

Links

Text available via DOI:

Keywords

View graph of relations

Deep learning-based aberration compensation improves contrast and resolution in fluorescence microscopy

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Published

Standard

Deep learning-based aberration compensation improves contrast and resolution in fluorescence microscopy. / Guo, M.; Wu, Y.; Hobson, C.M. et al.
In: Nature Communications, Vol. 16, No. 1, 313, 02.01.2025.

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Harvard

Guo, M, Wu, Y, Hobson, CM, Su, Y, Qian, S, Krueger, E, Christensen, R, Kroeschell, G, Bui, J, Chaw, M, Zhang, L, Liu, J, Hou, X, Han, X, Lu, Z, Ma, X, Zhovmer, A, Combs, C, Moyle, M, Yemini, E, Liu, H, Liu, Z, Benedetto, A, La Riviere, P, Colón-Ramos, D & Shroff, H 2025, 'Deep learning-based aberration compensation improves contrast and resolution in fluorescence microscopy', Nature Communications, vol. 16, no. 1, 313. https://doi.org/10.1038/s41467-024-55267-x

APA

Guo, M., Wu, Y., Hobson, C. M., Su, Y., Qian, S., Krueger, E., Christensen, R., Kroeschell, G., Bui, J., Chaw, M., Zhang, L., Liu, J., Hou, X., Han, X., Lu, Z., Ma, X., Zhovmer, A., Combs, C., Moyle, M., ... Shroff, H. (2025). Deep learning-based aberration compensation improves contrast and resolution in fluorescence microscopy. Nature Communications, 16(1), Article 313. https://doi.org/10.1038/s41467-024-55267-x

Vancouver

Guo M, Wu Y, Hobson CM, Su Y, Qian S, Krueger E et al. Deep learning-based aberration compensation improves contrast and resolution in fluorescence microscopy. Nature Communications. 2025 Jan 2;16(1):313. doi: 10.1038/s41467-024-55267-x

Author

Guo, M. ; Wu, Y. ; Hobson, C.M. et al. / Deep learning-based aberration compensation improves contrast and resolution in fluorescence microscopy. In: Nature Communications. 2025 ; Vol. 16, No. 1.

Bibtex

@article{4ece7c7b1a68436f837cd053e3b4374b,
title = "Deep learning-based aberration compensation improves contrast and resolution in fluorescence microscopy",
abstract = "Optical aberrations hinder fluorescence microscopy of thick samples, reducing image signal, contrast, and resolution. Here we introduce a deep learning-based strategy for aberration compensation, improving image quality without slowing image acquisition, applying additional dose, or introducing more optics. Our method (i) introduces synthetic aberrations to images acquired on the shallow side of image stacks, making them resemble those acquired deeper into the volume and (ii) trains neural networks to reverse the effect of these aberrations. We use simulations and experiments to show that applying the trained {\textquoteleft}de-aberration{\textquoteright} networks outperforms alternative methods, providing restoration on par with adaptive optics techniques; and subsequently apply the networks to diverse datasets captured with confocal, light-sheet, multi-photon, and super-resolution microscopy. In all cases, the improved quality of the restored data facilitates qualitative image inspection and improves downstream image quantitation, including orientational analysis of blood vessels in mouse tissue and improved membrane and nuclear segmentation in C. elegans embryos. ",
keywords = "blood, detection method, image analysis, image resolution, instrumentation, segmentation, signal processing, simulation, article, blood vessel, Caenorhabditis elegans, compensation, contrast, controlled study, deep learning, embryo, fluorescence, fluorescence microscopy, image quality, microscopy, mouse, nerve cell network, nonhuman, photon, animal, artificial neural network, confocal microscopy, image processing, procedures, Animals, Deep Learning, Image Processing, Computer-Assisted, Mice, Microscopy, Confocal, Microscopy, Fluorescence, Neural Networks, Computer",
author = "M. Guo and Y. Wu and C.M. Hobson and Y. Su and S. Qian and E. Krueger and R. Christensen and G. Kroeschell and J. Bui and M. Chaw and L. Zhang and J. Liu and X. Hou and X. Han and Z. Lu and X. Ma and A. Zhovmer and C. Combs and M. Moyle and E. Yemini and H. Liu and Z. Liu and A. Benedetto and {La Riviere}, P. and D. Col{\'o}n-Ramos and H. Shroff",
year = "2025",
month = jan,
day = "2",
doi = "10.1038/s41467-024-55267-x",
language = "English",
volume = "16",
journal = "Nature Communications",
issn = "2041-1723",
publisher = "Nature Publishing Group",
number = "1",

}

RIS

TY - JOUR

T1 - Deep learning-based aberration compensation improves contrast and resolution in fluorescence microscopy

AU - Guo, M.

AU - Wu, Y.

AU - Hobson, C.M.

AU - Su, Y.

AU - Qian, S.

AU - Krueger, E.

AU - Christensen, R.

AU - Kroeschell, G.

AU - Bui, J.

AU - Chaw, M.

AU - Zhang, L.

AU - Liu, J.

AU - Hou, X.

AU - Han, X.

AU - Lu, Z.

AU - Ma, X.

AU - Zhovmer, A.

AU - Combs, C.

AU - Moyle, M.

AU - Yemini, E.

AU - Liu, H.

AU - Liu, Z.

AU - Benedetto, A.

AU - La Riviere, P.

AU - Colón-Ramos, D.

AU - Shroff, H.

PY - 2025/1/2

Y1 - 2025/1/2

N2 - Optical aberrations hinder fluorescence microscopy of thick samples, reducing image signal, contrast, and resolution. Here we introduce a deep learning-based strategy for aberration compensation, improving image quality without slowing image acquisition, applying additional dose, or introducing more optics. Our method (i) introduces synthetic aberrations to images acquired on the shallow side of image stacks, making them resemble those acquired deeper into the volume and (ii) trains neural networks to reverse the effect of these aberrations. We use simulations and experiments to show that applying the trained ‘de-aberration’ networks outperforms alternative methods, providing restoration on par with adaptive optics techniques; and subsequently apply the networks to diverse datasets captured with confocal, light-sheet, multi-photon, and super-resolution microscopy. In all cases, the improved quality of the restored data facilitates qualitative image inspection and improves downstream image quantitation, including orientational analysis of blood vessels in mouse tissue and improved membrane and nuclear segmentation in C. elegans embryos.

AB - Optical aberrations hinder fluorescence microscopy of thick samples, reducing image signal, contrast, and resolution. Here we introduce a deep learning-based strategy for aberration compensation, improving image quality without slowing image acquisition, applying additional dose, or introducing more optics. Our method (i) introduces synthetic aberrations to images acquired on the shallow side of image stacks, making them resemble those acquired deeper into the volume and (ii) trains neural networks to reverse the effect of these aberrations. We use simulations and experiments to show that applying the trained ‘de-aberration’ networks outperforms alternative methods, providing restoration on par with adaptive optics techniques; and subsequently apply the networks to diverse datasets captured with confocal, light-sheet, multi-photon, and super-resolution microscopy. In all cases, the improved quality of the restored data facilitates qualitative image inspection and improves downstream image quantitation, including orientational analysis of blood vessels in mouse tissue and improved membrane and nuclear segmentation in C. elegans embryos.

KW - blood

KW - detection method

KW - image analysis

KW - image resolution

KW - instrumentation

KW - segmentation

KW - signal processing

KW - simulation

KW - article

KW - blood vessel

KW - Caenorhabditis elegans

KW - compensation

KW - contrast

KW - controlled study

KW - deep learning

KW - embryo

KW - fluorescence

KW - fluorescence microscopy

KW - image quality

KW - microscopy

KW - mouse

KW - nerve cell network

KW - nonhuman

KW - photon

KW - animal

KW - artificial neural network

KW - confocal microscopy

KW - image processing

KW - procedures

KW - Animals

KW - Deep Learning

KW - Image Processing, Computer-Assisted

KW - Mice

KW - Microscopy, Confocal

KW - Microscopy, Fluorescence

KW - Neural Networks, Computer

U2 - 10.1038/s41467-024-55267-x

DO - 10.1038/s41467-024-55267-x

M3 - Journal article

VL - 16

JO - Nature Communications

JF - Nature Communications

SN - 2041-1723

IS - 1

M1 - 313

ER -