Generalized method to design phase masks for 3D super-resolution microscopy.

Area of Science:

• Optics and Photonics
• Microscopy
• Computational Imaging

Background:

• Phase modulation is key for engineering point spread functions (PSFs) in 3D super-resolution microscopy.
• Current methods struggle to design phase masks for arbitrary PSFs and cannot simultaneously encode spatial and temporal data.
• A robust algorithm is needed to recover phase masks for desired PSF shapes.

Purpose of the Study:

• To introduce a generalized phase mask design method using optimization.
• To develop and compare algorithms for efficient phase mask recovery.
• To demonstrate simultaneous encoding of 3D spatial and temporal information using engineered PSFs.

Main Methods:

• Developed and compared stochastic gradient descent and Gauss-Newton algorithms for phase mask recovery.
• Utilized multi-level light lithography for in-house fabrication of phase masks.
• Experimentally validated a stretching-lobe phase mask for encoding spatio-temporal data.

Main Results:

• The Gauss-Newton algorithm demonstrated significantly faster convergence compared to stochastic gradient descent.
• A novel stretching-lobe phase mask was designed to simultaneously encode 3D spatial and temporal information.
• Phase masks were successfully fabricated and experimentally validated, including one encoding four letters at different depths.

Conclusions:

• The developed optimization algorithms provide a robust method for designing arbitrary point spread functions.
• The novel stretching-lobe phase mask enables simultaneous acquisition of 3D spatial and temporal data in microscopy.
• The generalized approach is applicable to various microscopy applications requiring complex PSF engineering.