Abstract

Optical Coherence Tomography (OCT) is a technique that allows imaging tissue in three spatial dimensions. Such a technique makes it possible to examine the subsurface of the tissue. The depth of penetration into the tissue can be tailored by tuning the wavelength of the light source. While in some cases it is desirable to obtain deep penetration of the sample, when scanning for cancerous changes, it may only be necessary to penetrate the first few hundred micrometres. The use of a shorter wavelength, while decreasing the penetration depth, will improve the resolution of the instrument. While images from OCT systems contain speckle and other artefacts, there are methods of evaluating the information by using image processing techniques. Of particular interest is the scattering coefficient that can be derived from the OCT data. Using discriminant techniques on the scattering data (such as principal components analysis), gives a sensitive way of differentiating between changes in structure in the tissue. An extensive data collection was performed on cervical tissue using samples that ranged from normal to invasive cancer. The histopathology of each sample was gathered and was classified from normal to cancer. The scattering profiles of the data were averaged and gradient analysis was performed, showing that for small distances into the sample there is a significant difference between scattering profiles between cancerous and normal tissue. PCA was also performed on the data showing grouping into various stages of cancer.

© 2011 OSA/SPIE