Measuring Tissue Dispersion Using the Cross-correlation of Half- Spectrum Optical Coherence Tomography Images

In Optical Coherence tomography (OCT), dispersion mismatches cause degradation of the image resolution and are, thus, compensated accordingly. However, dispersion is specific to the material that is causing the effect and can, therefore, carry useful information regarding the composition of the samples. In this summary, we propose a novel technique for estimating tissue dispersion by calculating the cross-correlation of images acquired at different center wavelengths to estimate the shift between their features, also known as walk-off, and use that to calculate the dispersion. Since a distinct reflector is not required, this method is applicable to any sample and can even be implemented in vivo and in situ in human tissues. The proposed technique was verified ex vivo resulting in Group Velocity Dispersion (GVD) values comparable to those obtained from estimating the walk-off from a mirror, as described in the literature. The applicability to cancer diagnosis was evaluated on a small set of gastrointestinal normal and cancer OCT images. Using the statistics of the GVD estimates, tissue classification resulted in 100% sensitivity and 81% specificity (92% correct classification rate). The success of these preliminary results indicates the potential of the proposed method, which should be further investigated to elucidate its advantages and limitations.Copyright 2019 Society of Photo‑Optical Instrumentation Engineers (SPIE). One print or electronic copy may be made for personal use only. Systematic reproduction and distribution, duplication of any material in this publication for a fee or for commercial purposes, and modification of the contents of the publication are prohibited.Christos Photiou, Evgenia Bousi, Ioanna Zouvani, Costas Pitris, "Measuring tissue dispersion using the cross-correlation of half-spectrum optical coherence tomography images," Proc. SPIE 10867, Optical Coherence Tomography and Coherence Domain Optical Methods in Biomedicine XXIII, 108670D (22 February 2019); doi: 10.1117/12.251094