Sparse deconvolution spatial light microscopy in two and three dimensions
Methods and a computer program product for applying deconvolution to spatial light interference microscopy for resolution enhancement with respect to the diffraction limit in two and three dimensions. By exploiting the sparsity properties of the phase images, which is prominent in many biological imaging applications, and modeling of the image formation via complex fields, the very fine structures can be recovered which were blurred by the optics. The resolution improvement leads to higher accuracy in monitoring dynamic activity over time. Experiments with primary brain cells, i.e. neurons and glial cells, reveal new subdiffraction structures and motions. This new information can be used for studying vesicle transport in neurons, which may shed light on dynamic cell functioning. Finally, the method may flexibly incorporate a wide range of image models for different applications and can be utilized for all imaging modalities acquiring complex field images.
1. A method for deriving a quantitative phase contrast image of a physical specimen, the method comprising:
a. illuminating the specimen with a beam characterized by a focus through a condenser annulus that introduces a phase delay that is an integral multiple of π/2;
b. temporally modulating at least one of the phase and amplitude of the spatial frequency components of the spatially transformed image to obtain interferograms at a plurality of relative phase offsets;
c. deriving a quantitative phase image of the specimen based on the interferograms, the image associating an intensity and a phase with each location in a field of view;
d. casting the quantitative phase image in terms of a sparse image model;
e. calculating a phase-preserving deconvolved complex image, deconvolved, in the sparse image model, with respect to an optical system degradation by iteratively maximizing a posterior probability distribution separately with respect to a complex field e [i{tilde over (Φ)}(x,y)] ; and
f. displaying the phase-preserving deconvolved complex image.
2. A method in accordance with claim 1 , further comprising a step of setting a constant field magnitude prior to casting the quantitate phase image in terms of a sparse image model.
3. A method in accordance with claim 2 , wherein the at least one linear transform includes a directional contourlet.
4. A method in accordance with claim 2 , further comprising modeling signal-independent noise via an independent Gaussian noise model with a single parameter.
5. A method in accordance with claim 2 , wherein the step of calculating a deconvolved complex image includes incorporating a regularization term into the posterior portability distribution.
6. A method in accordance with claim 2 , wherein the step of calculating a deconvolved complex image entails a linear maximization problem.
7. A method in accordance with claim 1 , further comprising applying at least one linear transform as a high-pass filter to the quantitative phase image.
8. A method in accordance with claim 1 , wherein the step of illuminating further comprises scanning the focus to a plurality of depths within the specimen.
9. A method in accordance with claim 8 , wherein the step of calculating a deconvolved complex image includes deconvolving the quantitative phase image to obtain a three-dimensional phase representation of the specimen at each of a plurality of distances relative to a fiducial plane.
10. A method in accordance with claim 8 , further comprising applying a plurality of linear transforms on all planes of the quantitative phase image.
11. A computer program product for use on a computer system for deriving a quantitative phase contrast image of a physical specimen, the computer program product comprising a computer usable tangible medium having computer readable program code fixed thereon, the computer readable program code including:
a. program code for inputting interferograms obtained by illumination of a specimen at a plurality of relative phase offsets;
b. program code for deriving a quantitative phase image of the specimen based on the interferograms;
c. program code for casting the quantitative phase image in terms of a sparse image model; and
d. program code for calculating a deconvolved complex image preserving quantitative phase information of the quantitative phase image, deconvolved, in the sparse image model, with respect to an optical system degradation, by iteratively maximizing a posterior probability distribution separately with respect to a complex field e [i{tilde over (Φ)}(x,y)] .