Multilayer Tissue-Like Optical Phantom; a Model for Skin in Optical Coherence Tomography Imaging - Mohammad R. N. Avanaki, Ali Hojjat, and Adrian G. Podeleanu

Optical coherence tomography (OCT) is an advanced high-resolution non-invasive imaging tool, which delivers three-dimensional (3D) images from the microstructure compartments within the skin tissue. Using OCT one can extract optical properties (scattering coefficient and anisotropy factor) of normal or diseased skin to generate an optical model for skin, which can be used for diagnosis. To verify and

validate the optical properties extraction algorithm a tissue-like optical phantom is needed. If the phantom can be multilayer, it can model the skin more accurately.

WHAT IS PHANTOM AND WHY IS IT NEEDED?

Phantom is a virtual tissue with well-controlled optical properties (refractive index, scattering coefficient, anisotropy factor and absorption coefficient) and can be constructed in solid or liquid states depending on its application. Phantoms are being used for testing the design of the system, system optimisation, and performance evaluation of the system. In optical imaging modalities in particular, phantoms are used for measurement/evaluation of the system parameters such as longitudinal and transversal resolutions, image contrast, point spread function (PSF), system sensitivity, system differentiability between two types of tissues, and system detectability of a certain concentration. In this study, we used phantoms to evaluate the accuracy and precision of the optical properties extraction algorithm in calculating the values of scattering coefficient and anisotropy factor.

HOW CAN WE EXTRACT OPTICAL PROPERTIES FROM OCT IMAGES?

The optical properties (scattering coefficient and anisotropy factor) can be extracted from the OCT images by fitting the OCT signal which is obtained based on extended Huygens-Fresnel (EHF) principle, onto the averaged A-line vector in a specific region. The fitting procedure and corresponding curves are shown in Figure 1 and Figure 2, respectively. One can generate an optical model for skin by specifying the range of variation of the optical properties for the healthy skin. Then the values out of this range can be used for the diagnosis of diseased skins.