Practical sensorless aberration estimation for 3D microscopy with deep learning

Saha, Debayan; Schmidt, Uwe; Zhang, Qinrong; Barbotin, Aurélien; Hu, Qi; Ji, Na; Booth, Martin J.; Weigert, Martin; Myers, Eugene W.

doi:10.1364/oe.401933

Cited by 43 publications

(33 citation statements)

References 31 publications

(47 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…According to the recently published papers by Saha et al, it can be confirmed that RMSE parameter is used as a quantitative index for the analysis of the utility of deep learning techniques in 3D microscopy images [ 31 ]. In addition, according to a study conducted at Linden et al, the RMSE evaluation parameter was used to confirm the difference between true and estimation data when analyzing the usefulness of the image super resolution algorithm for microscope images [ 32 ].…”

Section: Resultsmentioning

confidence: 99%

Blind Deconvolution Based on Compressed Sensing with bi-l0-l2-norm Regularization in Light Microscopy Image

Kim

2021

IJERPH

View full text Add to dashboard Cite

Blind deconvolution of light microscopy images could improve the ability of distinguishing cell-level substances. In this study, we investigated the blind deconvolution framework for a light microscope image, which combines the benefits of bi-l0-l2-norm regularization with compressed sensing and conjugated gradient algorithms. Several existing regularization approaches were limited by staircase artifacts (or cartooned artifacts) and noise amplification. Thus, we implemented our strategy to overcome these problems using the bi-l0-l2-norm regularization proposed. It was investigated through simulations and experiments using optical microscopy images including the background noise. The sharpness was improved through the successful image restoration while minimizing the noise amplification. In addition, quantitative factors of the restored images, including the intensity profile, root-mean-square error (RMSE), edge preservation index (EPI), structural similarity index measure (SSIM), and normalized noise power spectrum, were improved compared to those of existing or comparative images. In particular, the results of using the proposed method showed RMSE, EPI, and SSIM values of approximately 0.12, 0.81, and 0.88 when compared with the reference. In addition, RMSE, EPI, and SSIM values in the restored image were proven to be improved by about 5.97, 1.26, and 1.61 times compared with the degraded image. Consequently, the proposed method is expected to be effective for image restoration and to reduce the cost of a high-performance light microscope.

show abstract

Section: Resultsmentioning

confidence: 99%

Blind Deconvolution Based on Compressed Sensing with bi-l0-l2-norm Regularization in Light Microscopy Image

Kim

2021

IJERPH

View full text Add to dashboard Cite

show abstract

“…Recently, important advances have also been made to improve SPT and single-molecule SR imaging in densely labeled samples using deep learning (for a review, see e.g., Möckl et al, 2020a ). These developments include using neural nets to optimize and localize engineered PSFs faster and for overlapping emitters (Nehme et al, 2018 , 2020 ; Zhang et al, 2018 ), phase retrieval of aberrations, and background correction (Paine and Fienup, 2018 ; Möckl et al, 2019b , 2020b ; Saha et al, 2020 ). The next steps in this field are to standardize controls to ensure the validity of the analysis, as well as to make these approaches more easily accessible and comparable for a wide range of users.…”

Section: Discussionmentioning

confidence: 99%

Light Sheet Illumination for 3D Single-Molecule Super-Resolution Imaging of Neuronal Synapses

Gagliano

Nelson

Saliba

et al. 2021

Front. Synaptic Neurosci.

View full text Add to dashboard Cite

The function of the neuronal synapse depends on the dynamics and interactions of individual molecules at the nanoscale. With the development of single-molecule super-resolution microscopy over the last decades, researchers now have a powerful and versatile imaging tool for mapping the molecular mechanisms behind the biological function. However, imaging of thicker samples, such as mammalian cells and tissue, in all three dimensions is still challenging due to increased fluorescence background and imaging volumes. The combination of single-molecule imaging with light sheet illumination is an emerging approach that allows for imaging of biological samples with reduced fluorescence background, photobleaching, and photodamage. In this review, we first present a brief overview of light sheet illumination and previous super-resolution techniques used for imaging of neurons and synapses. We then provide an in-depth technical review of the fundamental concepts and the current state of the art in the fields of three-dimensional single-molecule tracking and super-resolution imaging with light sheet illumination. We review how light sheet illumination can improve single-molecule tracking and super-resolution imaging in individual neurons and synapses, and we discuss emerging perspectives and new innovations that have the potential to enable and improve single-molecule imaging in brain tissue.

show abstract

“…Wavefront-sensorless AO methods are commonly used because they require no extra hardware for direct wavefront sensing and, hence, allow for simpler optical designs and avoid non-common-path sensing errors. A range of wavefront-sensorless AO schemes exist, such as modal [ 9 ], pupil segmentation zonal [ 10 ], deep learning [ 11 ], and blind searching algorithm methods [ 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 ]. When based on the modal method, the correction speed of the system is fast, but the performance of the control method depends on an accurate mathematical model, and the range of correcting aberrations is limited.…”

Section: Introductionmentioning

confidence: 99%

Active Aberration Correction with Adaptive Coefficient SPGD Algorithm for Laser Scanning Confocal Microscope

KunHua

Zhang

et al. 2022

Sensors

View full text Add to dashboard Cite

A laser scanning confocal microscope (LSCM) is an effective scientific instrument for studying sub-micron structures, and it has been widely used in the field of biological detection. However, the illumination depth of LSCMs is limited due to the optical aberrations introduced by living biological tissue, which acts as an optical medium with a non-uniform refractive index, resulting in a significant dispersion of the focus of LSCM illumination light and, hence, a loss in the resolution of the image. In this study, to minimize the effect of optical aberrations, an image-based adaptive optics technology using an optimized stochastic parallel gradient descent (SPGD) algorithm with an adaptive coefficient is applied to the optical path of an LSCM system. The effectiveness of the proposed aberration correction approach is experimentally evaluated in the LSCM system. The results illustrate that the proposed adaptive optics system with an adaptive coefficient SPGD algorithm can effectively reduce the interference caused by aberrations during depth imaging.

show abstract

Practical sensorless aberration estimation for 3D microscopy with deep learning

Cited by 43 publications

References 31 publications

Blind Deconvolution Based on Compressed Sensing with bi-l0-l2-norm Regularization in Light Microscopy Image

Blind Deconvolution Based on Compressed Sensing with bi-l0-l2-norm Regularization in Light Microscopy Image

Light Sheet Illumination for 3D Single-Molecule Super-Resolution Imaging of Neuronal Synapses

Active Aberration Correction with Adaptive Coefficient SPGD Algorithm for Laser Scanning Confocal Microscope

Contact Info

Product

Resources

About