Texture analysis for tissue classification of optical coherence tomography images

Optical coherence tomography (OCT) is a cross-sectional imaging modality capable of acquiring images to depths of a few millimeters at resolutions ranging from 10-15 μm. This makes OCT useful for visualizing layers and structures within the tissue, but not effective for seeing in vivo cellular level detail. Random spatially dependent speckle patterns were seen in our images due to the coherent properties of light utilized in OCT. These speckle patterns are dependent on various optical parameters of the system, including numerical aperture, as well as the size and distribution of light scattering particles within the sample. The purpose of this study is to evaluate the application of statistical and spectral texture analysis techniques for differentiating tissue types based on the structural and speckle content in OCT images. Good correct classification rates were obtained when five different bovine tissues were compared in pairs, averaging 80% correct, and reasonable rates were obtained comparing normal vs. abnormal mouse lung tissue, averaging 64.0% and 88.6%, respectively. This study has shown that texture analysis of OCT images may be capable of differentiating tissue types without reliance on visible structures.