Three-dimensional structured illumination microscopy with enhanced axial resolution.

Area of Science:

• Biophysics
• Optical Microscopy
• Cell Biology

Background:

• Three-dimensional structured illumination microscopy (3D SIM) offers enhanced resolution over conventional microscopy.
• However, its axial resolution is typically limited to approximately 300 nm, restricting nanoscale investigations in the z-axis.

Purpose of the Study:

• To develop and present novel methods for improving axial resolution in 3D SIM.
• To achieve near-isotropic imaging with enhanced resolution for detailed cellular analysis.

Main Methods:

• Implementation of a four-beam interference technique using a mirror opposite the sample.
• Development of a deep learning-based approach for image reconstruction and resolution enhancement.
• Combination of deep learning with denoising for time-lapse volumetric imaging.

Main Results:

• Achieved near-isotropic imaging with approximately 120-nm lateral and 160-nm axial resolution using the mirror method.
• Attained approximately 120-nm isotropic resolution with the deep learning method.
• Demonstrated successful application in imaging vimentin, microtubules, caveolae, lysosomes, and T cell immune synapses.

Conclusions:

• The presented methods significantly improve axial resolution in 3D SIM with minimal optical system modifications.
• These advancements enable high-resolution, near-isotropic imaging of cellular structures and dynamics.
• The deep learning approach facilitates efficient volumetric and time-lapse imaging for advanced cell biology research.