3D Image Super-Resolution by Fluorophore Fluctuations and MA-TIRF Microscopy Reconstruction (3D-COL0RME)

Stergiopoulou, Vasiliki; Calatroni, Luca; Schaub, Sébastien; Blanc‐Féraud, Laure

doi:10.1109/isbi52829.2022.9761572

Cited by 3 publications

(2 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent works [32,30], the authors improved upon the artefacts created by SPARCOM on simulated and real noisy data by considering a non-convex continuous exact approximation of the 0 pseudo-norm previously employed in [27,9] and incorporating a further background and noise optimisation step within a two-step procedure allowing also for intensity estimation. Such approach, called COL0RME (COvariance-based 0 super-Resolution Microscopy with intensity Estimation), has been also applied to 3D image data in [31] showing high-quality reconstructions and high applicability in common scenarios. A major limitation of SPARCOM and COL0RME when applied to challenging real-word contexts is that, due to their mathematical structure, they tend to promote point-like structures also in the case of samples characterised by the presence of different structures (e.g., thin filaments).…”

Section: Introductionmentioning

confidence: 99%

Fluorescence image deconvolution microscopy via generative adversarial learning (FluoGAN)

Cachia

Stergiopoulou²,

Calatroni

et al. 2023

Inverse Problems

Self Cite

View full text Add to dashboard Cite

We propose FluoGAN, an unsupervised hybrid approach combining the physical modelling of fluorescence microscopy timelapse acquisitions with a generative adversarial learning procedure for the problem of image deconvolution. Differently from standard approaches combining a least-square data term based on one (long-time exposure) image with sparsity-promoting regularisation terms, FluoGAN relies on a data term being the distributional distance between the fluctuating observed timelapse (short-time exposure images) and the generative model. Such distance is computed by adversarial training of two competing architectures: a physics-inspired {\em generator} simulating the fluctuating behaviour as a Poisson process of the observed images combined with blur and undersampling, and a standard convolutional {\em discriminator} network. FluoGAN is a fully unsupervised approach requiring only a fluctuating sequence of blurred, undersampled and noisy images of the sample of interest as input. It can be complemented with prior knowledge on the desired solution such as sparsity, non-negativity etc. After having described the main ideas behind FluoGAN, we formulate the corresponding optimisation problem and report several results on simulated and real phantoms used by microscopy engineers to quantitatively assess spatial resolution. The comparison of FluoGAN with state-of-the-art methodologies shows improved resolution, allowing for high-precision reconstructions of fine structures in challenging real {\em Ostreopsis cf Ovata} data. The FluoGAN code is available at: https://github.com/cmayeul/FluoGAN.

show abstract

Section: Introductionmentioning

confidence: 99%

Fluorescence image deconvolution microscopy via generative adversarial learning (FluoGAN)

Cachia

Stergiopoulou²,

Calatroni

et al. 2023

Inverse Problems

Self Cite

View full text Add to dashboard Cite

show abstract

“…Some of these approaches are: the Super-resolution Optical Fluctuation Imaging (SOFI) approach [5] where the lack of correlation between distinct emitters is exploited by analyzing high-order statistics, the Super-Resolution Radial Fluctuations (SRRF) [11] microscopy, where super-resolution is achieved by calculating at each frame the degree of local symmetry and, finally, the Sparsity-based Super-resolution Correlation Microscopy (SPARCOM) [25] which models the sparse distribution of the fluorescent molecules via the use of a convex 1 regularisation applied on the emitters' covariance matrix. To improve the performance of SPARCOM, the Covariancebased 0 super-resolution microscopy with intensity estimation (COL0RME) method [27,29] has been proposed to estimate both molecule positions and intensities, which is a valuable piece of information in several applications, such as, e.g., 3D imaging [28], by means of a two-step procedure relying on hand-crafted sparsity promoting regularisers. The approach further estimates noise statistics and background terms (containing out-of-focus molecules).…”

Section: Introductionmentioning

confidence: 99%

Fluctuation-based deconvolution in fluorescence microscopy using plug-and-play denoisers

Stergiopoulou¹,

Mukherjee²,

Calatroni³

et al. 2023

Preprint

View full text Add to dashboard Cite

The spatial resolution of images of living samples obtained by fluorescence microscopes is physically limited due to the diffraction of visible light, which makes the study of entities of size less than the diffraction barrier (around 200 nm in the x-y plane) very challenging. To overcome this limitation, several deconvolution and super-resolution techniques have been proposed. Within the framework of inverse problems, modern approaches in fluorescence microscopy reconstruct a superresolved image from a temporal stack of frames by carefully designing suitable hand-crafted sparsity-promoting regularisers. Numerically, such approaches are solved by proximal gradient-based iterative schemes. Aiming at obtaining a reconstruction more adapted to sample geometries (e.g. thin filaments), we adopt a plug-and-play denoising approach with convergence guarantees and replace the proximity operator associated with the explicit image regulariser with an image denoiser (i.e. a pretrained network) which, upon appropriate training, mimics the action of an implicit prior. To account for the independence of the fluctuations between molecules, the model relies on second-order statistics. The denoiser is then trained on covariance images coming from data representing sequences of fluctuating fluorescent molecules with filament structure. The method is evaluated on both simulated and real fluorescence microscopy images, showing its ability to correctly reconstruct filament structures with high values of peak signal-to-noise ratio (PSNR).

show abstract

Fluctuation-Based Deconvolution in Fluorescence Microscopy Using Plug-and-Play Denoisers

Stergiopoulou

Mukherjee

Calatroni

et al. 2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

3D Image Super-Resolution by Fluorophore Fluctuations and MA-TIRF Microscopy Reconstruction (3D-COL0RME)

Cited by 3 publications

References 13 publications

Fluorescence image deconvolution microscopy via generative adversarial learning (FluoGAN)

Fluorescence image deconvolution microscopy via generative adversarial learning (FluoGAN)

Fluctuation-based deconvolution in fluorescence microscopy using plug-and-play denoisers

Fluctuation-Based Deconvolution in Fluorescence Microscopy Using Plug-and-Play Denoisers

Contact Info

Product

Resources

About