Phantom and computational studies towards the clinical translation of gas in scattering media absorption spectroscopy into neonatal respiratory care

Abstract

Everything except vacuum is heterogeneous to some extent. Even media that we consider homogeneous (such as pure gases and water) can be taken apart into individual heterogeneities (such as atoms and molecules), which can be distinguished with a sufficiently fine probe.

Absorption spectroscopy was extensively used by Robert Bunsen and Gustav Kirchhoff in the 1860’s to separate, identify and measure various chemical substances. They defined a line of research, where traces of elements were just detectable with the aid of specialized instruments like the spectroscope, and since then, the absorption lines have been subject of experimental and theoretical developments. Today, we know that the nature of the absorption lines can be described by quantum mechanical changes induced in the atoms or molecules, and with the advances in light sources and sensing technologies, absorption spectroscopy has become a tremendously useful tool with a wide range of applications.

The studies presented in this thesis are related to gas absorption spectroscopy, in particular, a technique called GASMAS, which stands for “GAs in Scattering Media Absorption Spectroscopy”. This spectroscopy technique was introduced in Lund University in 2001 by S. Svanberg’s group, to study the spectral features of gases inside porous or hollowed scattering media, combining laser spectroscopy with sensitive modulation techniques. Unlike solids and liquids, which have a smooth absorption and scattering wavelength dependence (1 − 10 nm), gases exhibit sharp absorptive features (10−4 nm). This difference between the absorption spectra of solid state matter and gases, is the corner stone of this technique. In a typical GASMAS measurement, the laser wavelength is scanned across at least one of the absorption lines of the gas of interest. The small gas absorption signal (embedded in the scattered spectrum from the bulk material) is then filtered from the detected signal, making it possible to retrieve the gas concentrations and study their diffusion dynamics using the principles of the Beer-Lambert law. Although there is evidence of the potential of GASMAS to sense oxygen and water vapor in human cavities, such as the ear, nasal sinuses, lungs, intestines and hip bone, one the most promising clinical applications could be the lung function assessment in neonates. The focus of this thesis is to investigate the potential of translating GASMAS into such an application, combining a computational and experimental approach. Most of the work was done in a collaboration between Biophotonics@Tyndall, the Infant Centre (hosted at the Cork University Maternity Hospital-UCC) and the Swedish industry partner, GPX Medical who have built a pioneering GASMAS instrument, suitable for clinical use. The motivation behind this collaborative work, is to assist clinicians in the monitoring of lung function in premature newborns, as their lungs lack structural and biochemical maturation, which can result in respiratory failure.

Currently, the use of GASMAS is limited to observational studies with healthy babies. Thus, the improvement and optimization of the technology depends on feasibility tests with tissue-like models. Phantoms mimicking the geometry and optical properties of the main thoracic organs, were created to study the influence of source detector positioning and chest physiognomy in the GASMAS signals. A functional phantom resembling the anatomy, temperature and humidity of the respiratory zone, was also developed to investigate the potential of GASMAS technique in measuring changes in inflated volume.

The optimization of source-detector configurations over the thorax is one of the challenges in the clinical translation of GASMAS. It is crucial to define the optimal probe positioning, to obtain the highest possible signal reaching the detector, which also carries information of the gas absorption in lung tissue. Computational studies are then used to simulate the light transport in accurate anthropomorphic models, which contributes with the understanding of near infra-red interaction with the thorax, and most of all, to find the probe locations for which the detection of gas absorption is feasible, and enhance the data acquisition in future clinical studies.

This document includes the theoretical background of GASMAS, the basics of respiratory physiology, and the current methods for clinical monitoring and diagnostics of lung pathologies in neonates. The following two chapters, show how the developed phantom and computational models enable the recreation of different clinical scenarios, suitable for GASMAS studies. The main contribute is the identification of the minimum requirements necessary to further improve and advance towards a GASMAS bedside clinical device, that can potentially be used, for lung function assessment and monitoring in neonatal respiratory health.