More about HKUST
Graph-based 3D Vascular Segmentation and Its Applications
The Hong Kong University of Science and Technology Department of Computer Science and Engineering PhD Thesis Defence Title: "Graph-based 3D Vascular Segmentation and Its Applications" By Miss Ning ZHU Abstract In recent years, vessel disease, which is one of the major causes of death around the world, has become an important health problem. Vessel segmentation is an important technique, and it can help the diagnosis, visualization, treatment and surgery planning for vessel diseases. In this thesis, we first review existing techniques for 3D vessel segmentation. Then several vessel segmentation techniques including centerline detection method (Minimum Average-cost Path for 4D Optimization), centerline-based segmentation methods (Optimal Cross Sections and Graph Optimization via graph cuts) and segmentation methods without ROI restriction (Random Walks with ACF, Random Walk based CEDA Optimization, Power-watershed based Optimization with TMT, Single Target Segmentation with AP-TMT) are proposed. Among these methods, Random Walk based CEDA Optimization has been evaluated on public evaluation system and demonstrated to be more accurate than all state-of-the-art methods on segmentation accuracy for healthy vessels. As for diseased vessel parts, comparable results are obtained. Vessel stenosis is one major type of vascular diseases. Because of the complex structure of stenoses, both detection and quantification of stenoses are challenging. In this thesis, two segmentation and feature based methods (PWIS and type-based) and one learning based stenosis detection method is proposed and evaluated to be effective on improving the stenosis detection rates. Date: Tuesday, 19 August 2014 Time: 10:00am - 12:00noon Venue: Room 4472 Lifts 25/26 Chairman: Prof. Ying Chau (CBME) Committee Members: Prof. Albert Chung (Supervisor) Prof. Chiew-Lan Tai Prof. Dit-Yan Yeung Prof. Weichuan Yu (ECE) Prof. Pheng-Ann Heng (Comp. Sci. & Engg., CUHK) **** ALL are Welcome ****