Chair for Computer Aided Medical Procedures & Augmented Reality
Lehrstuhl für Informatikanwendungen in der Medizin & Augmented Reality

THIS WEBPAGE IS DEPRECATED - please visit our new website

Diploma thesis

[Subtitle] Real-Time Respiratory Motion Tracking: Roadmap Correction for Endovascular Interventions
Thesis by: Selen Atasoy
Advisor: Nassir Navab
Supervision by: Martin Groher
in collaboration with Siemens angiography and X-ray
Due date: 15.06.2007


Over the past years endovascular interventions became gold standard for treatment of many diseases such as cancer or blockage of blood flow. These interventions are performed percutaneously using small instruments such as catheters. While navigating the catheter through the blood vessels the interventional radiologist relies on 2D X-ray fluoroscopy images, which display the catheter and parts of the patient anatomy in real-time. However, the visualization of blood vessels is only possible for a short-period of time when contrast agent is injected. In order to overcome this limitation 3D roadmaps, which are acquired by rotational angiography, can be overlayed on live 2D fluoroscopy images. Thus, the amount of contrast agent can be minimized. Moreover, the procedure time can be reduced by providing a significant help in catheter navigation. However, in abdominal interventions this technique needs to be extended by an appropriate respiratory motion correction. In this thesis a method for real-time respiratory motion correction for 3D static roadmaps is proposed with the objective of introducing corrected 3D roadmaps into clinical workflow for abdominal interventions. The proposed method consists of two steps. First, a filtering is applied to the 2D fluoroscopy images to enhance the catheter and reduce the noise level. Then, a part of the catheter is tracked in the improved images using templatematching in order to estimate the catheter motion induced by patient’s respiration. In order to deal with possible catheter deformations, an adequate template update strategy is proposed. Real-time implementations of these steps allow an intraoperative application of the algorithm. The accuracy and robustness of the algorithm are demonstrated with extensive experimental studies using both simulated image sequences and clinical image sequences.



Title: Real-Time Respiratory Motion Tracking: Roadmap Correction for Endovascular Interventions
Student: Selen Atasoy
Director: Nassir Navab
Supervisor: Martin Groher
Type: DA/MA/BA
Status: finished
Start: 2007/01/15
Finish: 2007/06/15

Edit | Attach | Refresh | Diffs | More | Revision r1.3 - 28 Nov 2008 - 10:36 - MarcusToennis