Our long-term goal is to combine state-of-the-art, three-dimensional (3D) image guidance with a new method of vascular segmentation and description in order to increase the speed and reduce the risk of difficult endovascular procedures. A specific example of a difficult procedure is transjugular intrahepatic portosystemic shunt formation (TIPS), during which the clinician inserts a needle through a patient's liver in order to connect one venous circulation with another. The target cannot be seen during liver puncture, thus leading to multiple needle passes and high patient risk. Our goal is to provide the clinician accurate, 3D image guidance in order to increase the speed and to decrease the risks of the procedure. Methods include: 1) Segmentation of the vasculature and liver from preoperatively acquired MR or CT data, 2) Intraoperative 3D-2D registration of the segmented 3D vessels with intraoperatively acquired biplane digital subtraction angiograms (DSA), 3) Development of fast, real-time tracking methods to detect changes in the position of the target portal vein resulting from patient respiration or needle pressure on the liver during the procedure, 4) 3D reconstruction of the TIPS needle from biplane views, thus placing the needle within the coordinate system of the 3D vasculature, 5) Integration of information and visualization by means of stereo, head-tracked display, and 6) A four-phase evaluation process. The methods developed under this proposal should be of immediate benefit to TIPS. Moreover, these methods are generally applicable to a wide variety of endovascular procedures, and are thus immediately relevant to the hundreds of thousands of patients with tumors or vascular disease that undergo endovascular treatment each year. Indeed, the methods to be developed under this proposal have the potential to change how endovascular procedures are performed throughout the body.
| Piliere, Guillaume; Van Horn, Mark H; Dixon, Robert et al. (2009) Vessel target location estimation during the TIPS procedure. Med Image Anal 13:519-29 |
| Jomier, Julien; Bullitt, Elizabeth; Van Horn, Mark et al. (2006) 3D/2D model-to-image registration applied to TIPS surgery. Med Image Comput Comput Assist Interv 9:662-9 |
| Lorenzen, Peter; Prastawa, Marcel; Davis, Brad et al. (2006) Multi-modal image set registration and atlas formation. Med Image Anal 10:440-51 |
| Bullitt, Elizabeth; Muller, Keith E; Jung, Inkyung et al. (2005) Analyzing attributes of vessel populations. Med Image Anal 9:39-49 |
| Lorenzen, Peter; Davis, Brad; Joshi, Sarang (2005) Unbiased atlas formation via large deformations metric mapping. Med Image Comput Comput Assist Interv 8:411-8 |
| Maupu, Damien; Van Horn, Mark H; Weeks, Susan et al. (2005) 3D stereo interactive medical visualization. IEEE Comput Graph Appl 25:67-71 |
| Pathak, Chetna; Van Horn, Mark; Weeks, Susan et al. (2005) Comparison of simultaneous and sequential two-view registration for 3D/2D registration of vascular images. Med Image Comput Comput Assist Interv 8:239-46 |
| Prastawa, Marcel; Bullitt, Elizabeth; Ho, Sean et al. (2004) A brain tumor segmentation framework based on outlier detection. Med Image Anal 8:275-83 |
| Fridman, Yonatan; Pizer, Stephen M; Aylward, Stephen et al. (2004) Extracting branching tubular object geometry via cores. Med Image Anal 8:169-76 |
