Advanced

Diffuse Reflectance Spectroscopy: Using a Monte Carlo method to determine chromophore compositions of tissue

Kobayashi Frisk, Lisa LU (2016) FYSM60 20161
Atomic Physics
Department of Physics
Abstract
In this work, inverse diffusion equation and Monte Carlo methods are used in conjunction with diffuse reflectance spectroscopy to develop a protocol for evaluating the chromophore compositions of tissue mimicking liquid phantoms. A novel two spectrometer system is used which provides real time calibration of any intensity fluctuations of the light source. The calibration of this setup is investigated in detail. In order to analyse the limitations of the two theoretical models (diffusion equation and Monte Carlo), two optical probes with source-detector separations of 0.5 and 2 mm are used. The final evaluated fittings of whole blood, deoxygenated haemoglobin, oxygenated haemoglobin concentrations and scattering coefficient values are... (More)
In this work, inverse diffusion equation and Monte Carlo methods are used in conjunction with diffuse reflectance spectroscopy to develop a protocol for evaluating the chromophore compositions of tissue mimicking liquid phantoms. A novel two spectrometer system is used which provides real time calibration of any intensity fluctuations of the light source. The calibration of this setup is investigated in detail. In order to analyse the limitations of the two theoretical models (diffusion equation and Monte Carlo), two optical probes with source-detector separations of 0.5 and 2 mm are used. The final evaluated fittings of whole blood, deoxygenated haemoglobin, oxygenated haemoglobin concentrations and scattering coefficient values are relatively good. Possible approaches for improving the experimental data are discussed. The Monte Carlo method is shown to give better results than the diffusion equation for short probe designs, as expected from theory. However, for longer probe designs, the diffusion theory evaluates the data better. This is likely due to worse planar fitting of the Monte Carlo look-up table at long source-detector separations. In general however, the results are promising and indicate good prospects for using this system on real biological systems. (Less)
Popular Abstract
8.2 million people around the world died of cancer in 2014, and in Sweden cancer rates have been increasing steadily for the past 20 years. Since cancer is such a common disease, cancer research is a very `hot' topic. Researchers are working hard not only to develop more efficient treatments, but also to develop better diagnostic techniques. In this work, light is used to develop a diagnostic method that could be used during cancer treatment.

The anatomy of tumour tissue changes when the tumour grows, and also while the tumour is being treated. More specifically, the amount of blood and oxygen in the tumour tissue changes. If the amount of blood and oxygen in tissue could be determined accurately, this would provide invaluable... (More)
8.2 million people around the world died of cancer in 2014, and in Sweden cancer rates have been increasing steadily for the past 20 years. Since cancer is such a common disease, cancer research is a very `hot' topic. Researchers are working hard not only to develop more efficient treatments, but also to develop better diagnostic techniques. In this work, light is used to develop a diagnostic method that could be used during cancer treatment.

The anatomy of tumour tissue changes when the tumour grows, and also while the tumour is being treated. More specifically, the amount of blood and oxygen in the tumour tissue changes. If the amount of blood and oxygen in tissue could be determined accurately, this would provide invaluable information when treating tumours. Light is able to be used to determine the amount of blood or oxygen in tissue because the way light travels through tissue depends on what is inside of it. In order to understand how light can be used to determine the composition of tissue one can consider runners running a course. One runner runs up a mountain at high speed and the other runs a flat course at low speed. If we see the runners after a certain distance, we can determine which course each runner took by analysing their heart rates, amount of sweat produced, etc. The runners here are analogous to light, and the course they travel is analogous to tissue. If we shine light onto a tissue and then detect light that exits the tissue a certain distance away, we can determine what was in that tissue by observing how the spectrum of the light changed after propagating through it. This is the basic principle of the technique used in this work, known as Diffuse Reflectance Spectroscopy.

In order to use Diffuse Reflectance Spectroscopy to determine the composition of tissue, one must be able to theoretically model the propagation of light through tissue. One model, known as the diffusion equation, is easy to implement and commonly used. However, the diffusion equation cannot correctly model all tissue types and experimental setups. Another model, known as Monte Carlo, is more difficult to implement but is less limiting than the diffusion equation. In this work, both the diffusion equation and Monte Carlo are used to try to extract the composition of solutions (referred to as liquid phantoms) that mimick the optical properties of biological tissue. The results, despite some errors, indicate a good outlook for using Diffuse Reflectance Spectroscopy as a diagnostic method during cancer therapy. (Less)
Please use this url to cite or link to this publication:
author
Kobayashi Frisk, Lisa LU
supervisor
organization
course
FYSM60 20161
year
type
H2 - Master's Degree (Two Years)
subject
keywords
Photonics, Biophotonics, Diffuse Reflectance, Spectroscopy, Monte Carlo, Diffusion
language
English
id
8882554
date added to LUP
2016-08-10 11:54:36
date last changed
2016-08-10 11:54:36
@misc{8882554,
  abstract     = {In this work, inverse diffusion equation and Monte Carlo methods are used in conjunction with diffuse reflectance spectroscopy to develop a protocol for evaluating the chromophore compositions of tissue mimicking liquid phantoms. A novel two spectrometer system is used which provides real time calibration of any intensity fluctuations of the light source. The calibration of this setup is investigated in detail. In order to analyse the limitations of the two theoretical models (diffusion equation and Monte Carlo), two optical probes with source-detector separations of 0.5 and 2 mm are used. The final evaluated fittings of whole blood, deoxygenated haemoglobin, oxygenated haemoglobin concentrations and scattering coefficient values are relatively good. Possible approaches for improving the experimental data are discussed. The Monte Carlo method is shown to give better results than the diffusion equation for short probe designs, as expected from theory. However, for longer probe designs, the diffusion theory evaluates the data better. This is likely due to worse planar fitting of the Monte Carlo look-up table at long source-detector separations. In general however, the results are promising and indicate good prospects for using this system on real biological systems.},
  author       = {Kobayashi Frisk, Lisa},
  keyword      = {Photonics,Biophotonics,Diffuse Reflectance,Spectroscopy,Monte Carlo,Diffusion},
  language     = {eng},
  note         = {Student Paper},
  title        = {Diffuse Reflectance Spectroscopy: Using a Monte Carlo method to determine chromophore compositions of tissue},
  year         = {2016},
}