Simulating optical memory effects and the scanning of foci using wavefront shaping in tissue-like scattering media

Jun, 2023·
Jake Bewick
Jake Bewick
,
Peter Munro
,
Simon Arridge
,
James Guggenheim
· 2 min read
Abstract
Wavefront shaping could enable focussing light deep inside scattering media, increasing the depth and resolution of imaging techniques like optical microscopy and optical coherence tomography. However, factors like rapid decorrelation times due to microscale motion and thermal variation make focusing in living tissue difficult. A way to ease the requirements could be exploiting prior information provided by memory effects. For example, this might allow partially or wholly scanning a focus. To study this and related ideas, a computational model was developed to simulate the generation and correlations of foci formed by WFS in scattering media. Predictions of the angular memory range were consistent with experimental observations. Furthermore, correlations observed between optical phase maps required to focus at different positions suggested correlation-based priors might enable accelerated focussing. This work could pave the way to faster optical focussing and thus deeper imaging in living tissue.
Type
Publication
European Conference on Biomedical Optics 2023
A manuscript and the acompanying presentation for a talk given at the European Conference on Biomedical Optics in Munich.
Simulation summary of wavefront shaping
Figure: The angular memory effect causes correlations in a speckle pattern as the incident light is tilted.

Biological tissue strongly scatters visible and near-infrared light, resulting in a significant reduction in intensity at depth. As a result, imaging modalities such as optical coherence tomography or photoacoustic microscopy are constrained to image superficial tissues. One method to extend the penetration depth of light in these modalities is wavefront shaping (WFS) - a technique in which the incident light is spatially modulated to control its propagation through scattering media, potentially allowing light to be focused at arbitrary locations within tissue.

Computational methods for simulating WFS could complement experimental investigation, which is often constrained by a lack of control. For example, with computational approaches it becomes possible to evaluate the field inside a given medium, resolve both amplitude and phase information, and fully manipulate the optical properties and geometry of said medium. Unfortunately, current approaches to simulate light propagation through biological tissue are either too computationally intensive to model volumes large enough to significantly benefit from WFS, or too incomplete to model underlying deterministic scattering and interference processes accurately.

We propose coupling the T-matrix method with the discrete particle model to create an efficient but rigorous simulation of light propagation through biological tissue. The T-matrix method works by propagating light through a medium of scattering particles such that the total field is a superposition of the scattered fields associated with each sphere, creating a solvable matrix system. By controlling the density, radii, and refractive indices of these spheres we are able to design bespoke domains with desired optical properties potentially matching those of tissue.

To simulate the complex beams found in WFS we apply the angular spectrum method whereby arbitrary beams can be decomposed into a spectrum of plane waves incident at varying angles which can be simulated sequentially. In this manuscript we demonstrate how our T-matrix discrete particle approach coupled with an angular spectrum decomposition of the incident field can be used to efficiently simulate optical foci generation through turbid media. We begin by replicating the original Vellekoop and Mosk WFS experiment by simulating the generation of an optical focus through a titanium dioxide scattering layer. We then simulate focus generation through a tissue-like 100µm3 sample volume.