3D deconvolution of spherically aberrated images using commercial software

Area of Science:

• Microscopy
• Optical Physics
• Biophysics

Background:

• Refractive index mismatch between specimens and immersion oil causes spherical aberration.
• Spherical aberration distorts the point spread function and axial coordinates in 3D fluorescence imaging.
• Accurate 3D image restoration requires accounting for these optical aberrations.

Purpose of the Study:

• To investigate the accuracy of 3D shape restoration using commercial software (Huygens, Autoquant).
• To evaluate the impact of refractive index mismatch and aberration correction on 3D image restoration.
• To assess the axial scaling accuracy in 3D wide-field fluorescence microscopy.

Main Methods:

• Utilized well-defined specimens (E. coli, actin filaments) with known refractive indices.
• Applied commercial deconvolution software (Huygens, Autoquant) with and without aberration correction.
• Measured axial scaling factors and compared lateral and axial dimensions of restored objects.

Main Results:

• Aberration correction significantly reduced axial blur compared to uncorrected deconvolution.
• Complete recovery of cylindrical symmetry for bacteria and actin filaments was not achieved; axial width remained 3-5 times larger than lateral.
• Restoration quality showed moderate sensitivity to specimen refractive index; reduced objective numerical aperture (NA) sometimes improved results.
• Axial scaling was accurately described by the paraxial formula, even with high-NA objectives.

Conclusions:

• Commercial software effectively reduces axial blur due to refractive index mismatch but has limitations in fully restoring specimen symmetry.
• Accurate axial scaling is achievable with standard formulas, crucial for reliable 3D image analysis.
• Careful consideration of optical parameters and software settings is necessary for precise 3D fluorescence image restoration.