Abstract:
Since its invention in the 1990s optical coherence tomography (OCT) has
evolved from a promising optical imaging modality to a well-established and powerful
modality with a broad spectrum of applications. Cross-sectional images are directly
formed by an interferometric measurement of the answer of objects to the radiation by
light. These images then provide detailed information about the internal microstruc-
ture of biological tissues which is used for clinical investigation and diagnosis.
The increasing interest in this modality over the past has shifted the focus to
quantitative methods which are used to gain additional information, in the form of
physical properties, which is hidden in the OCT data. This field of research is referred
to as quantitative optical coherence tomography (QOCT).
The main achievement of this dissertation was the quantification of optical pa-
rameters from experimental OCT data which also forms the core of this presentation.
Mathematically, we formulate the quantification in OCT as an inverse scattering prob-
lem based on Maxwell’s equation system.
In the first part of this presentation, the direct scattering problem is discussed.
Locally, we reduce it to an one-dimensional electromagnetic wave scattering problem.
There, the object of interest is assumed to show a layered structure, a realistic scenario
for biological samples imaged by OCT systems, which is expressed by a piecewise
constant function (with respect to depth) as the optical parameter. The presented
problem, herein, is based on a Gaussian beam model for the strongly focused laser
light illumination within the OCT system. While the commonly applied plane wave
model failed to be accurate enough, such a Gaussian beam model has proven to
predict precisely the effects visible in an OCT system. We discuss experiments for
the calibration of necessary features for this model, like the focus position within the
imaging system. Finally, we verify by numerical experiments that the simulations on
the basis of this model match with the experimental data.
In the second part, we discuss the corresponding inverse problem where we pro-
pose a layer-by-layer reconstruction method based on this Gaussian beam model. The
quantification problem, herein, is recast for each step as a minimization problem be-
tween the forward prediction and the (experimental) data. We give a characterization
of uniqueness and show the applicability of the method by reconstructing the refractive
indices from both simulated and experimental OCT data. There the reconstructed
parameters match with the available ground truth.

Zoom-Link:

Meeting ID: 632 7750 8291
Passcode: 716656