Texture Analysis of Optical-Coherence Tomographic Images

Student: Kirk W. Gossage

Optical coherence tomography (OCT) acquires cross-sectional images of tissue by measuring back-reflected light. Images from in vivo OCT systems typically have a resolution of 10-15mm, and are thus best suited for visualizing features in the tens to hundreds of microns size range such as tissue layers or glands. Many normal and abnormal tissues lack features in this size range so it may appear that OCT is unsuitable for identification of these tissues. However, examination of feature-poor OCT images reveals that they frequently display a characteristic repetitive structure due to speckle.

The purpose of this study was 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. Experiments yielded promising classification rates, demonstrating that texture analysis of OCT images may be capable of differentiating tissue types in the absence of visibly identifiable structures.

Here are some example OCT images of mouse fat and lung, respectively:

Principal dissertation advisor: Prof. Jennifer K. Barton, Biomedical Engineering.