Computational Approach to Dark-Field Optical Diffraction Tomography
Abstract
The measurement of three-dimensional (3D) images and the analysis of subcellular organelles are crucial for the study of the pathophysiology of cells and tissues. Optical diffraction tomography (ODT) facilitates label-free and quantitative imaging of live cells by reconstructing 3D refractive index (RI) distributions. In many cases, however, the contrast in RI distributions is not strong enough to effectively distinguish subcellular organelles in live cells. To realize label-free and quantitative imaging of subcellular organelles in unlabeled live cells with enhanced contrasts, we present a computational approach using ODT. We demonstrate that the contrast of ODT can be enhanced via spatial high-pass filtering in a 3D spatial frequency domain, and that it yields theoretically equivalent results to physical dark-field illumination. Without changing the optical instruments used in ODT, subcellular organelles in live cells are clearly distinguished by applying a simple but effective computational approach that is validated by comparison with 3D epifluorescence images. We expect that the proposed method will satisfy the demand for label-free organelle observations, and will be extended to fully utilize complex information in 3D RI distributions.
Cite
@article{arxiv.1910.02773,
title = {Computational Approach to Dark-Field Optical Diffraction Tomography},
author = {Taean Chang and Seungwoo Shin and Moosung Lee and YongKeun Park},
journal= {arXiv preprint arXiv:1910.02773},
year = {2019}
}