Comparison of different metrics for analysis and visualization in spectroscopic optical coherence tomography

Abstract

Spectroscopic Optical Coherence Tomography (S-OCT) extracts depth resolved spectra that are inherently available from OCT signals. The back scattered spectra contain useful functional information regarding the sample, since the light is altered by wavelength dependent absorption and scattering caused by chromophores and structures of the sample. Two aspects dominate the performance of S-OCT: (1) the spectral analysis processing method used to obtain the spatially-resolved spectroscopic information and (2) the metrics used to visualize and interpret relevant sample features. In this work, we focus on the second aspect, where we will compare established and novel metrics for S-OCT. These concepts include the adaptation of methods known from multispectral imaging and modern signal processing approaches such as pattern recognition. To compare the performance of the metrics in a quantitative manner, we use phantoms with microsphere scatterers of different sizes that are below the system's resolution and therefore cannot be differentiated using intensity based OCT images. We show that the analysis of the spectral features can clearly separate areas with different scattering properties in multi-layer phantoms. Finally, we demonstrate the performance of our approach for contrast enhancement in bovine articular cartilage.

Spectral domain optical coherence tomography for ex vivo brain tumor analysis

“…Afterwards Principal Component Analysis was employed for feature reduction. The first five principal components were taken for k-means clustering [14]. After the data points were grouped to one of the five clusters, a color map was generated.…”

“…Here the spectroscopic analysis is performed first with a narrow and then with a broad window and both results are combined by multiplication [11]. The dual window method as well as pattern recognition and machine learning algorithms were used in this study to enhance contrast between different images based on spectroscopic features [12][13] [14]. …”

Spectroscopic Low Coherence Interferometry

“…A more comprehensive and validated comparison of the time-frequency analysis methods for SOCT and LCS described in this chapter is described in Refs [49][50][51]. Other analysis methods for the visualization of spectroscopic information in SOCT images are compared in Ref [52]. …”

Correlation of the derivative as a robust estimator of scatterer size in optical coherence tomography (OCT) [Invited]

“…Variations of the above techniques were adopted in other studies. For example, Oldenburg et al used the autocorrelation width of backscattered spectra, at 80% of the peak value, in order to enhance the contrast of OCT images of macrophages and fibroblasts [21], whereas Kartakoullis et al and Jaedicke et al combined the spectral information with principal component analysis (PCA) and clustering algorithms with the intention of differentiating phantom samples consisting of microspheres with different diameters [23,24]. In addition, Tay et al suggested that the use of multiple bandwidths can improve the sensitivity of scatterer size estimates, demonstrating the technique in spectroscopic OCT images where solutions of 0.5 and 45 μm microspheres could be clearly distinguished [25].…”