DVDeconv: An Open-Source MATLAB Toolbox for Depth-Variant Asymmetric Deconvolution of Fluorescence Micrographs

Area of Science:

• Biomedical imaging
• Microscopy
• Computational imaging

Background:

• 3D microscopy images are crucial for cellular structure investigation.
• Diffraction limitations cause depth-variant and asymmetric blur in 3D microscopy images.
• Existing deconvolution methods and datasets fail to address depth-variant asymmetric blur.

Purpose of the Study:

• To develop an open-source solution for depth-variant asymmetric deconvolution in microscopy.
• To provide a dataset for evaluating deconvolution algorithms under realistic conditions.

Main Methods:

• Developed DVDeconv, a MATLAB-based software with a graphical interface.
• Included a dataset, a depth-variant asymmetric point spread function generator, and deconvolution algorithms.
• Validated the method through experimental results.

Main Results:

• DVDeconv accurately removes depth-variant asymmetric blur.
• The developed dataset effectively reflects actual imaging conditions.
• Experimental results demonstrate the efficacy of the deconvolution algorithms.

Conclusions:

• DVDeconv provides an effective solution for depth-variant asymmetric deconvolution in microscopy.
• The DVDeconv dataset serves as a valuable resource for benchmarking future deconvolution techniques.
• This work advances the accuracy of 3D cellular imaging and analysis.