Speckle image properties in optical coherence tomography

Most identification of tissues in OCT images has relied on the presence or absence of features and layers. However, in some pathologies as well as some normal tissues OCT images appear homogeneous. Examination of these images reveals that they display a characteristic repetitive structure due to speckle. Since speckle is influenced by the local index of refraction mismatches, it may be possible to differentiate between different types of tissues based on analysis of the speckle pattern. The determination of tissue type may be supported as well by local contrast distribution analysis or speckle decorrelation degree, which are widely used in measurement and characterization of surface roughness. In this study we examined three areas: 1) the application of speckle theory based on surface roughness to a three-dimensional media and a short coherence length light source, 2) the effect that the optical system design has on the received speckle distribution, and the optimum optical system geometry for speckle analysis, and 3) the speckle properties of OCT images of tissue phantoms and various tissues such as fat and muscle. Results obtained from two methods of speckle analysis (texture analysis and speckle contrast) were compared for their ability to differentiate between tissue types.