Wolf phase tomography (WPT) of transparent structures using partially coherent illumination

TL;DR

Wolf phase tomography (WPT) is a novel method that uses partially coherent light to directly reconstruct 3D refractive index distributions of transparent structures without Fourier transforms, enabling high-throughput imaging of live cells.

Contribution

This paper introduces Wolf phase tomography (WPT), a new diffraction tomography technique leveraging partially coherent fields for direct 3D imaging without Fourier transforms.

Findings

Successfully reconstructs 3D RI distributions from SLIM data.

Validates WPT with microbeads, spermatozoa, and neural networks.

Achieves RI sensitivity of approximately 10^-5.

Abstract

Diffraction tomography using coherent holographic imaging has been proposed by Emil Wolf in 1969 to extract 3D information from transparent, inhomogeneous objects. At the same time, the Wolf equations describe the propagation correlations associated with partially coherent fields. Combining these two concepts, here we present Wolf phase tomography (WPT), which is a method for performing diffraction tomography using partially coherent fields. The WPT reconstruction works in the direct space-time domain, without the need of Fourier transformation, and decouples the refractive index distribution from the thickness of the sample. We demonstrate the WPT principle using data acquired by spatial light interference microscopy (SLIM). SLIM is a quantitative phase imaging method that upgrades an existing phase contrast microscope by introducing controlled phase shifts between the incident and scattered fields. The illumination field in SLIM is spatially partially coherent (emerging from a ring-shaped pupil function) and of low temporal coherence (white light), thus, suitable for the Wolf equations. From three intensity measurements corresponding to different phase-contrast frames in SLIM, the 3D refractive index distribution is obtained right away by computing the Laplacian and second time derivative of the measured complex correlation function. The high-throughput and simplicity of this method enables the study of 3D, dynamic events in living cells across entire multi-well plates, with RI sensitivity on the order of 10-5. We validate WPT with measurements of standard samples (microbeads), spermatozoa, and live neural networks.