Wavefront shaping approaches for spectral domain optical coherence tomography

Abstract

Optical coherence tomography (OCT) enables sub-surface three dimensional imaging with micrometer resolution. The technique is based on the time-of-flight gated detection of light which is backscattered from a sample and has applications in non-destructive testing, metrology and contact-less and non-invasive medical diagnostics. With scattering media such as the human skin, the penetration depth is limited to just a few millimetres, on the other hand, and OCT imaging hence allows to investigate superficial sample layers only.

Scattering of light is a deterministic process. As a consequence, manipulation of the beam incident to a turbid sample yields control over the scattered field. Following this approach, a number of groups demonstrated iterative wavefront optimization algorithms to be able to focus light transmitted through or backscattered from opaque media. First applications to optical coherence tomography were shown to extend the penetration depth as well as to improve the signal-to-noise ratio when imaging biological tissue.

This work explores practical approaches to combine wavefront shaping techniques with OCT imaging. To this end, a compact spectral domain (SD-) OCT design is developed which enables single-pass and independent wavefront control at the reference and at sample beam. Iterative optimization of the phase pattern applied to the sample beam is shown to selectively enhance the amplitude of the OCT signal received from scattering media. In a more sophisticated approach, the acquisition of the time-resolved reflection matrix, which yields the linear dependence of the OCT signal on the field at the sample beam, is demonstrated. Subsequent wavefront optimization based on a phase conjugation algorithm is shown to enhance the OCT signal but not image artefacts, even though no attempt is made to actively suppress these artefacts. The approach is comparable to iterative wavefront optimization but yields a substantially improved acquisition speed. First imaging applications demonstrate the algorithm to enhance the signal-to-noise ratio and the penetration depth with scattering media, such as biological tissue, and to reduce the observed speckle contrast, similar to compounding algorithms. Furthermore, the acquisition of the reflection matrix and subsequent signal enhancement based on binary amplitude-only (on/off) beam shaping is presented for the first time. The technique can be implemented with digital micromirror devices which enable high-speed implementations.

The presented techniques constitute substantial improvements compared to previous works and yield promising results in the context of depth-enhanced OCT imaging with scattering biological tissue. Approaches to further enhance the performance and the acquisition speed for real-time in-vivo imaging applications are discussed.