Hybrid vascular tree extraction from angiography images /
استخراج شجرة الاوعية الدموية للصور الاشعاعية
Sara Saeed Mohamed Mohamed Abdeldayem ; Supervised Yasser Kadah , Inas A. Yassine
- Cairo : Sara Saeed Mohamed Mohamed Abdeldayem , 2015
- 84 P. : charts , facsimiles ; 30cm
Thesis (M.Sc.) - Cairo University - Faculty of Engineering - Department of System and Biomedical Engineering
We introduce a fast, simple, and an automatic hybrid approach, which extracts the vascular structures. First, we introduce the hybrid filter that merges a vessel enhancement and a bifurcation detectors filters. Second, we integrate this hybrid filter with a digital wave propagation system by mapping the filtered image to a cost image. A set of 40 fundus retinal images were used to study the enhancement gained by the hybrid filter and the wave propagation compared to the vessel filters. The results of comparison show an enhancement up to 10%,15%, and 50% in the dice coefficient, sensitivity and the response at bifurcation locations respectively