Image restoration for three-dimensional fluorescence microscopy using an orthonormal basis for efficient representation of depth-variant point-spread functions

Abstract

A depth-variant (DV) image restoration algorithm for wide field fluorescence microscopy, using an orthonormal basis decomposition of DV point-spread functions (PSFs), is investigated in this study. The efficient PSF representation is based on a previously developed principal component analysis (PCA), which is computationally intensive. We present an approach developed to reduce the number of DV PSFs required for the PCA computation, thereby making the PCA-based approach computationally tractable for thick samples. Restoration results from both synthetic and experimental images show consistency and that the proposed algorithm addresses efficiently depth-induced aberration using a small number of principal components. Comparison of the PCA-based algorithm with a previously-developed strata-based DV restoration algorithm demonstrates that the proposed method improves performance by 50% in terms of accuracy and simultaneously reduces the processing time by 64% using comparable computational resources.

Performance evaluation of extended depth of field microscopy in the presence of spherical aberration and noise

“…7 These methods have the advantage that they maintain the signal-to-noise properties of a conventional clear circular aperture (CCA) microscope; however, these methods do not directly address the degradation of the PSF at depth due to SA 8 and require 3-D postprocessing algorithms, which result in a longer postprocessing burden (in both time and computational resources) than EDFM methods. 9,10 In this paper, the performance of EDFM with selected phase mask designs (proposed as a result of our previous study) 11 and their robustness to SA and noise are evaluated in terms of image restoration quality. We present restored images that investigate the feasibility of SA-insensitive EDFM.…”

Experimental validation of a customized phase mask designed to enable efficient computational optical sectioning microscopy through wavefront encoding

“…3D imaging methods suitable for live specimen investigation should be non-invasive and should provide fast data acquisition. Although these two features are achieved in computational optical sectioning microscopy (COSM), which facilitates highresolution 3D imaging with a traditional wide-field fluorescence microscope [1][2][3], challenges due to the depth variability of the point spread function (PSF) [4] have resulted in the use of computationally intensive 3D depth-variant (DV) image reconstruction algorithms for quantitative imaging [5][6][7][8][9]. Because DV algorithms use multiple DV PSFs defined over the imaging volume, they require higher computational resources than space-invariant (SI), a.k.a.…”

“…Because DV algorithms use multiple DV PSFs defined over the imaging volume, they require higher computational resources than space-invariant (SI), a.k.a. deconvolution, algorithms [8].…”

3D Image Restoration Using Multiple Space-Varying Point Spread Functions

“…Results from simulated and experimental data that evaluate the three restoration methods [1][2][3] will also been shown.…”

“…We have been developing restoration methods that address this aberration using different mathematical approximations of the SV imaging model [1][2][3]. A recent extension of the point-spread function (PSF) model to predict more accurately sample-induced aberration [4] and the use of an orthonormal basis [2] for the efficient representation of the resulting SV PSFs have enable the development of a new restoration method that accounts for space-variant imaging [3], thereby extending our earlier methods that account only for depth-variant imaging [1,2]. Tradeoffs between accuracy and complexity as well as the needs of an imaging application guide the restoration method choice.…”