My short one semester stay at VUB-TONA was indeed one of the most fruitful learning experience in my life
A good cell identification is important to be able to differentiate between cells in a sample. In order to know the cell composition and to understand various cellular processes, a recognition of biomolecules inside living cells is necessary. For example in the field of tumor diagnosis and therapy monitoring of cancer patients, a reliable identification of tumor cells from a blood sample is indispensable .
Raman spectroscopy is a powerful technique for the analysis of such biological or chemical samples. Raman scattering refers to an inelastic light scattering from a sample. Although biomolecules have a low Raman cross section and hence require a long integration time, Raman spectra have a rich information content about the molecular constituents of the sample. This is done without labeling or without destroying the molecules. Therefore Raman spectroscopy is an appropriate technique for identification of cells, when using it in combination with statistical methods.
Since we want to evolve to `Lab-on-a-chip' (LOC) systems, an incorporation of Raman spectroscopy on microfluidic devices is necessary. LOC systems are systems that integrate several chemical processes on a chip of a few square cm (figure 1a). These miniaturized systems have been manufactured for several decennia. They consume reduced reagent and provide a fast sample analysis.
Figure 1: (a) ‘Lab-on-a-chip’ system, (b) Image of Chinese hamster ovary cell with probe locations, (c) Raman spectra from three different positions within a CHO cell: nucleus, cytoplasm and membrane .
The challenge of this master thesis is the detection of the Raman signal on the microfluidic device, because of the weak Raman signal and the background contamination. Nowadays the detection of the signals is still partly done with the use of a microscope objective. Measured Raman spectra are predominated by the background signals. When probing an ovary cell of a Chinese hamster (see figure 1b) the Raman features are difficult to discern due to the strong background (see figure 1c). This can for example be due to the presence of fluorescence. Different methods can be explored to suppress the background and to improve the signal-to-noise ratio of the measured Raman signal to perform a reliable identification of the cell under test.
 U. Neugebauer et al., Clinical and Biomedical Spectroscopy and Imaging II 2011.
 A. C. De Luca, M. Mazilu, A. Riches, C. S. Herrington, and K. Dholakia, Anal. Chem. 82, 738 (2010)
In this master thesis you will explore the possibilities to work towards totally integrated Raman spectroscopy on a microfluidic device. In a first step you will model -using raytracing software- a suitable design to detect Raman signals without the need for an external objective. This is possible by combining an optimum design with different possibilities to suppress the unwanted background and by using appropriate statistical methods. In a next step you will fabricate the polymer lab-on-a-chip and demonstrate its proof-of-concept and as such contribute to the challenging field of biophotonics.
|Diane De Coster||Supervisor|