MMF-PSFGenerator

Spatially variant point spread function (PSF) is a critical phenomenon in point-scanning-based multimode fiber (MMF) imaging, often causing image blur—particularly near the edges of the field of view (FOV) and at longer working distances. To achieve clear fluorescence images, spatially variant deconvolution is therefore essential. However, most existing approaches rely on experimentally measured 2D PSFs, which change with the imaging depth (z-plane) and require new measurements for each plane. This repository provides simulation codes to generate theoretical 3D spatially variant PSFs, with customizable parameters reflecting the imaging conditions. This simulation-based approach may offer a new perspective for performing deconvolution in MMF imaging.

├── MMF_PSF_Generator.m                # Function to simulate single 2D PSF
├── MMF_3DPSF_Generator.m              # Function to simulate single 3D PSF
├── Main_Spatial_Variant_PSF.m         # Function to simulate spatially variant 2D PSF        
├── MMF_Resolution_simulation.m        # Function to simulate theoretical resolution in 3D and fluorescence collection efficiency
├── PSFgrid.tif                        # Example spatially variant 2D PSF generated 
└── README.md                          # Documentation (this file)

Acknowledgment

Thanks Dr. Zhong Wen for helping derive the theory for this.

Contact

For any questions or comments about this code, please contact zdong@caltech.edu.