Vladlena Gorbunova from the eScience Center defended her PhD
Vlandlena Gorbunova from the eScience Center defended her PhD "Image Registration of Lung CT Scans for Monitoring Emphysema Progression" at DIKU on November 1th 2010
Chronic Obstructive Pulmonary Disease (COPD) is one of the main causes of death and disability; it is predicted to be the third leading cause of death worldwide after ischemic heart disease and cerebra-vascular disease in 2030. Longitudinal studies of COPD can help to discriminate between rapidly declining and stable subjects and provide necessary treatment on earlier stages of disease. In a longitudinal study of COPD, the conventional densitometry of lung CT scans provides a continuous measurement of disease progression, however the CT scores of emphysema are global measures quantifying the disease in the complete lung.
For the detailed regional analysis of longitudinal changes in lungs, one needs an accurate spatial correspondence between the sequential CT scans. In this thesis, several automatic image registration algorithms that establish a point-to-point correspondence between lung CT scans taken at different time points were developed and evaluated. Both intensity information and distinctive features visible in lung CT scans, e.g. blood vessels, were considered in the image registration. An important application of such image registration methods is the monitoring of emphysema progression in longitudinal studies. For this application, an algorithm to quantify changes in the distribution of image intensity and of local image features in corresponding regions was developed and evaluated.
Phone: (+45) 35 32 14 33
Office: E-Science 04.0.15B
For an electronic copy of the thesis, please contact Dina Riis Johannessen, firstname.lastname@example.org