From sample structure to optical properties and back: A theoretical framework for quantitative OCT and its clinical application

The aim of this thesis is to provide a methodology to extract quantitative parameters from optical coherence tomography (OCT) images and to relate these parameters to sample properties such as structure, organization and flow.
OCT uses near-infrared light to acquire 3D images of biological tissue, with a resolution around 5-15 μm and an imaging depth of approximately 2 mm. OCT can be a useful clinical tool for minimally-invasive, high-resolution imaging of tissue structures and morphology. In addition to morphological imaging, quantification of scattering properties from OCT images is a potential complementary source of information on microscopic tissue properties, which can be used clinically for tissue characterization. Quantitative parameters can be extracted from OCT data using an appropriate model for the OCT signal. However, the extracted parameters strongly depend on the applied methodology, and are sensitive to the selected input parameters. In order to establish the clinical value of quantitative OCT, the extracted parameters should be reliable and robust. Furthermore, the sensitivity of these parameters to clinically relevant changes in tissue should be determined.
In the first part of this thesis a model describing the OCT signal and the relation of quantitative OCT-parameters to sample properties of discrete random media is derived and validated. Furthermore, a robust method to extract quantitative parameters from OCT data is presented. The second part of this thesis covers two in vivo pilot studies on the feasibility of quantitative OCT during surgery, in which the obtained quantitative OCT-parameters are related to structural tissue properties and flow.