Scatterer size-based analysis of optical coherence tomography images

The early stages of malignancy, in most tissues, are characterized by unique cellular changes. Currently, these early changes are detectable only by confocal or multi-photon microscopy. Unfortunately, neither of the two imaging techniques can penetrate deep enough into the tissue to investigate the borders of thick lesions. A technique which would allow extraction of information regarding scatterer size from Optical Coherence Tomography (OCT) signals could prove a very powerful diagnostic tool and produce significant diagnostic insight. Such a procedure is proposed here. It is shown to be very effective in differentiating spectral differences which depend on scatterer size. The analysis of the OCT signal is based on spectral estimation techniques and statistical analysis. First, using autoregressive spectral estimation, it was deduced that tissues with different size scatterers exhibit marked differences in spectral content. Further, advanced analysis techniques, such as Principal Component Analysis (PCA) and Multivariate Analysis of Variance (MANOVA), provided more insight into the spectral changes. These techniques where tested on solutions of known scatterers and multilayered samples. The initial results are very encouraging and indicate that the spectral content of OCT signals can be used to extract scatterer size information. This technique can result in an extremely valuable tool for the investigation of disease tissue features which now remain below the resolution of OCT.