Scientific Focuses

Polarimetry is emerging as a promising, minimally invasive diagnostics tool.
The interaction of polarized light with matter often reveals features that are invisible to ordinary imaging techniques / diffuse optics. We have developed a versatile polarimetric microscope, that can measure with high speed cross- and auto-correlations
between arbitrarily polarized incident and backscattered light. This microscope has undegone a rigorous calibration process by performing measurements on well-characterized colloidal suspensions and comparing the results with Monte Carlo simulations.

Long term goals include differentiating tumorous from healthy tissues (in collaboration with the Neurosurgical Institute of the Inselspital Bern), diagnosing the severeness of burn injuries, as well as studying the fundamentals of polarized light propagation in random media, whereby more theoretical work is carried out.

We have built a confocal laser scanning microscope to perform fluorescence lifetime imaging to work on tissue diagnostics. The setup uses time-correlated single photon counting (TCSPC), together with a tunable picosecond pulsed excitation source.

Ongoing investigations include a collaboration with the Department of Chemistry and Biochemistry at the University of Bern, whereby the interactions of photosensitizers with their micro-environments are studied in the frame of photodynamic therapy.

Cilia are roughly 10_m long hair-like protrusions of the cell membrane. In the airways of higher organisms, their coordinated oscillatory beating movements causes directed motion of the mucosal fluid, cleaning the airways from adhering inhaled particles and harmful substances.

It is obvious that in order to achieve efficient directed transport, the ciliary motions must be, at least locally coordinated. Immotile cilia or lack of coordination are made responsible for different severe diseases, such as primary ciliary dyskinesia, therefore there is a demand for reliable diagnostic techniques. The key to every diagnosis is a sound knowledge of the healthy system. To achieve this goal, we focus on two complementary lines of action:

• We have developed a setup for the investigation and characterization of various aspects and parameters of mucociliary transport in an inter-species comparative study involving cow, sheep, pig, rabbit and turkey. In a collaboration with the clinic for zoo animals, exotic pets and wildlife (Univ. of Zurich), we have investigated the mucociliary clearance in snakes infected with boid inclusion body disease (BIBD). In BIBD-a_ected snakes amphophilic perinuclear inclusion bodies are found in the respiratory epithelium (see Figure), whose consequence to the mucociliary function has not been evaluated before. As pneumonia is commonly seen in BIBD-infected snakes, BIBD is suspected to impair mucociliary clearance in the respiratory tract.
   
• We have developed a theoretical model for explaining the self-organization that leads to mucociliary transport, based on interpreting ciliated cells as oscillating boolean actuators and their interactions as logical functions. Interactions between cells are triggered by virtual mucus lumps, which establish the network’s topology. This simple model exhibits the emergence of spatio-temporal patterns, that are accompanied by self-organized transport. The co-evolution of the network’s state and its topology allows to understand the mucociliary dynamics in the context of adaptive networks. Based on the study of a range of various model parameters, our main conclusion is that unciliated cells may improve the robustness of the spatio-temporal organization on ciliated epithelia, as they introduce a degree of modularity into the network’s topology. This provides a reasonable understanding of the observed patch-work character in the tracheal mucociliary system, which may be the expression of a modular mucocilairy network.

We have developed a general purpose Monte Carlo simulation package based on the concepts outlined by Rička et al. in “Optical-Thermal Response of Laser-Irradiated Tissue” (Springer, 2011) and initially intended to model light propagation in soft condensed matter such as biological tissues, but whose scope is far from being restricted to radiative transfer in tissues. The software is playfully named jaMCp3: just another Monte Carlo program for polarized photon propagation and composed of a user interface in IDL/GDL and a stand-alone photon path generation routine implemented in C++. Besides the Mie scattering model, the program includes a novel scattering model (the “polarized version of the generalized Henyey-Greenstein” scattering law) containing a structure factor parameter. The latter is particularly well-suited to describe light propagation in soft condensed matter such as biological tissue. Moreover, the simulations are conducted in a voxel space, where custom and complex geometrical structures can easily be generated. These also allow for the treatment of Fresnel reflection/refraction processes.

JaMCp3 has previously been subject to preliminary writings in its embryonic stages: “Simulating light propagation: towards realistic tissue models” (SPIE proceeding, 2011) and “Polarized Light Propagation in Biological Tissue: Towards Realistic Modeling” (Doctoral thesis, 2011). Yet, over the past years, jaMCp3 underwent rigorous testing stages (see references below). Certainly, the validation of such a MC program is a prime challenge, as quantitative comparisons, either with experiments or other simulation programs, are difficult to perform, and furthermore, the modeling of complex geometrical structures is bound to induce various numerical singularities, which might not be apparent at first glance. We are currently preparing a manuscript that details our program's main features and gives examples of its diverse possible outputs (3D fluence, 3D absorbed dose, polarimetric images...). The package, together with its user's manual, will soon be made available here for download, however, it is already possible to use it by contacting Günhan Akarçay ( hidayet.akarcay at iap.unibe.ch ).

References:
(1) Akarçay et al., “Determining the optical properties of a gelatin‑TiO2 phantom at 780 nm,” Biomed. Opt. Exp. 3 (2012).
(2) Hiltpold, “Bestimmung optischer Gewebeparameter,” Bachelor thesis (2012).
(3) Akarçay et al., “Monte Carlo modeling of polarized light propagation: Stokes vs Jones–Part I,” Appl. Opt. (2014).
(4) Akarçay et al., “Monte Carlo modeling of polarized light propagation: Stokes vs Jones–Part II,” Appl. Opt. (2014).
ERRATUM in (4): We noticed a small error in the last column (input state |O>) of the ''Primary data'' shown in Figure 1. The images corresponding to < L+ | and < L- | analyzers should be exchanged with the images obtained with < C- | and < C+ | analyzers, respectively. This was just a small bug in the representation of the data and does not affect the other figures/results. (See entirety of figures in the .pdf file linked at the bottom of this page.)

Optoacoustic (OA) Microscopy is a promising non-invasive medical imaging technique that is able to obtain 3D structural information by making use of the optoacoustic effect, i.e. the creation of ultrasonic pressure waves upon absorption of light. Combining this with the fact that the hemoglobin contained in red blood cells is the dominant absorber inside biological tissue, it provides high contrast 3D imaging of the vasculature at a spatial resolution superior to what can be achieved by MRI or micro-CT. Since blood perfusion and -oxygenation are important markers for cancer detection, OA microscopy could prove valuable in this context. Using OA microscopy, our group has studied the brain vasculature of malaria-infected mice (see Figure) and is engaged in the general development of the technique.

Spectral OA imaging exploits the wavelength-dependent optical absorption properties of specific chromophores in tissue (e.g. hemoglobin), with as goal to provide quantitative estimates of their spatially varying concentrations. A physiologically important example is the determination of local blood oxygen saturation, based on the distinct absorption spectra of oxy- and deoxyhemoglobin in the near-infrared range. This is of particular relevance for the study of oxygenation heterogeneity in tumors, the early detection and monitoring of cerebral ischemia in brain, and of other abnormalities characterized by a change in tissue oxygenation or perfusion status.

A central issue, which makes quantitative oxygenation measurements challenging, is the unknown extent of wavelength-dependent optical attenuation inside biological tissue. It engenders a spectral distortion of the recorded OA signals relative to the absorption spectrum of the chromophore's of interest. Only when this distortion is taken into account (which is referred to as "spectral correction"), a recovery of accurate quantitative information is possible.

As part of our research on quantitative deep OA imaging, we investigate techniques to correct for the spectral distortion of the recorded OA signals.

For clinical combined optoacoustic (OA) and ultrasound (US) imaging, maximum flexibility is provided when the irradiation optics is integrated with the ultrasound detector into a handheld probe. The drawback of this epi-illumination geometry is that the high laser intensity near the detection aperture generates strong background signals that clutter the OA image, thus limiting imaging depth (see Figure, phantom study). Our niche is the development of techniques for reducing this clutter with the goal to enable deep OA imaging. One of these techniques is localized vibration tagging (LOVIT): a long pulsed ultrasound beam generates acoustic radiation force (ARF) that induces localized tissue displacement at its focus. Subtraction of OA images acquired before and after the ARF push preserves true OA signal in the displacement focus while eliminating the clutter background (see Figure).