High space-bandwidth in quantitative phase imaging using partially spatially coherent digital holographic microscopy...

Area of Science:

• Biophotonics and Imaging
• Computational Microscopy
• Cellular Morphology

Background:

• Quantitative phase microscopy (QPM) offers label-free monitoring of subcellular morphology.
• QPM system performance relies on light source coherence and objective numerical aperture (NA).

Purpose of the Study:

• To develop a high space-bandwidth quantitative phase imaging technique.
• To enhance spatial sensitivity and resolution in QPM using partially spatially coherent light and deep learning.

Main Methods:

• Partially spatially coherent digital holographic microscopy (PSC-DHM) was employed.
• A generative adversarial network (GAN) was trained with low-resolution (LR) and high-resolution (HR) datasets from PSC-DHM.
• The method was tested on human red blood cells (RBC) and macrophages.

Main Results:

• The PSC-DHM + GAN approach improved the space-bandwidth product by 9× for both RBC and macrophage datasets.
• The network successfully predicted HR phase images from low NA lens captures.

Conclusions:

• The proposed PSC-DHM + GAN method significantly enhances quantitative phase imaging resolution.
• This approach holds potential for label-free tissue imaging, disease classification, and high-resolution tomography.