Title: Semi-automatic quantitative assessment of cancer-cell invasion 'in vitro' : an image-processing approach
Author: Hagglund, Samuel
Awarding Body: Kingston University
Current Institution: Kingston University
Date of Award: 2009
Availability of Full Text:
Access through EThOS:
Access through Institution:
EThOS Persistent ID: uk.bl.ethos.506729 
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
Abstract:
In western countries at least one third of the population develops cancer. The main cause of death in cancer patients is metastasis and there is no effective treatment for this complication. The situation can be improved by a better understanding of the cancer invasion process. In order to re­ veal new aspects of this dynamic process, a novel image-processing-based direct viewing cancer-cell invasion assay was developed and used with inverted wide-field microscopy. The combination of high-resolution 3D image-processing approaches with a custom-made flow chamber system enabled the quantification of the sarcoma-cell invasion process through a monolayer of endothelial cells in vitro. The image processing entailed the separation of positive cell signal from background noise and blur, which are inherent in 3D wide-field microscopy. The preparation and cell signal segmentation of wide-field images prior to quantification featured stochastic as well as deterministic techniques. The stochastic approach was based on a Gaussian Mixture Model to separate noise and background signal characteristics from positive cell signal which performed well in conditions with high signal-to-noise ratios. The. deterministic segmentation approach was based- on linear diffusion and performed well despite low signal-to-noise ratios as it assessed the diffusion rates of cell signal over multiple convolutions. The image-processing-based assay included the definition of two new parameters to quantify the invasion: Relative Invasion (RI) and Opening Rate of the Endothelial Monolayer (O REM). The first parameter RI measured the invasion as the percentage of sarcoma cell signal below the reconstructed monolayer surface. The second parameter O RE M evaluated the speed at which the sarcoma cells disassemble the monolayer in their strive to exit the flow channel. This assay was applied to metastatic rat sarcoma cells where the cells invaded monolayers of rat endothelial cells. After adhesion, the sarcoma cells initially invaded significantly faster under flow conditions compared to situations without shear stress. Later, however, the rate of invasion underflow decreased and the sarcoma cells without shear stress achieved significantly higher levels of invasion. These observations thus revealed the non-linear modulation of a tumour-cell invasion process by shear flow, demonstrating that tumour cells can respond to flow by enhancement of invasiveness in a similar way to white blood cells. In summary, the newly developed direct viewing assay provides a quantitative image-processing-based approach to assessing cancer invasion dynamics, which should lead to a better understanding of the mechanisms involved in cancer invasion and metastasis.
Keywords: Cancer studies : Computer science and informatics
Share: