Functional imaging with MHz-OCT

Abstract

Functional imaging enables the visualization of tissue structure alongside a map of its dynamic activity. This could include movements at the cellular level, mechanical changes, or metabolically induced activity patterns. Such information complements conventional imaging and can be particularly important where morphological differences alone do not allow for reliable diagnostic conclusions. In many clinical situations, this kind of functional additional information is crucial, for example, when assessing tissue function and integrity or localizing tumor tissue during surgery. The goal of current developments is therefore to provide imaging methods that can capture these functional signals quickly, non-invasively, and without the use of exogenous contrast agents, ideally in real-time and with a high resolution. Optical coherence tomography (OCT) is a non-invasive optical imaging technique that provides high-resolution cross-sectional images of biological tissue. In recent years, OCT has advanced considerably beyond its initial role as a purely structural imaging method. Notably, progress in achieving A-scan rates in the megahertz range has created new opportunities. With these so-called MHz-OCT systems, large data sets can be collected quickly and with high phase stability, which is essential for accurately visualizing the functional properties of tissue. In this work, two complementary functional approaches based on MHz-OCT were examined and further developed with a focus on their potential applications in biomedical imaging. In the first part of the work, the suitability of MHz-OCT for optical coherence elastography (OCE) to distinguish between healthy and tumorous brain tissue was evaluated. Two strategies were implemented for this purpose. One was a phase-sensitive method for the precise measurement of microscopic tissue displacements, and the other was an alternative contrast approach based on a Fourier analysis of the complex OCT signal. The approaches were experimentally validated using phantoms, ex vivo porcine brain tissue, and ex vivo human brain tumor tissue. The resulting elasticity maps were correlated with histological reference data and clearly showed differences in mechanical tissue stiffness. Additionally, analyses of stability and reproducibility were conducted to confirm the reliability and robustness of the method. In the second part of this work, dynamic MHz-OCT (dOCT) was used to study speckle‑based intensity fluctuations in ex vivo kidney tissue. The aim was to determine whether these signals could provide an additional imaging contrast that may be valuable in the context of kidney transplantation. To enable large-area imaging, a self-built, high‑precision, motorized three-axis linear robot was realized and comprehensively characterized in terms of its positioning accuracy. Furthermore, various temporal sampling strategies were investigated, with particular emphasis on non-equidistant sampling schemes aimed at reducing the number of repetitive frames while maintaining high dynamic contrast quality. Principal component analysis was used as part of the evaluation. The methods developed and analyzed in this dissertation demonstrate the strong potential of MHz-OCT for functional tissue analysis. Both OCE and dOCT offer a significantly deeper understanding of tissue composition and activity without relying on external contrast agents, tissue markers, or time-consuming histological processing. The concept shown in this work illustrates how dynamic MHz-OCT could support fast and objective tissue analysis in clinical practice. Additionally, the elastographic brain tissue measurements indicate that MHz-OCT may provide meaningful support in situations where existing intraoperative imaging techniques are insufficient.