Quantitative Phase Imaging

Quantitative phase imaging (QPI) is a label-free microscopy technique that measures the optical path length variations through a sample, providing quantitative information about its refractive index and thickness. Current research heavily utilizes deep learning, particularly diffusion models and convolutional neural networks, to improve phase reconstruction speed and accuracy, often addressing challenges like speckle noise in fiber endoscopy and aberration correction in traditional microscopy. These advancements are enabling faster, more robust QPI for diverse applications, including biomedical imaging (e.g., cell segmentation, pathology), materials science, and potentially even on-chip sensing through diffractive optical processors.