Meeting Banner
Abstract #0694

Artery-Vein Segmentation in Non-Contrast-Enhanced Flow-Independent 3D Peripheral Angiography

Serena Y. Yeung1, Kie Tae Kwon1, Bob S. Hu1, 2, Dwight G. Nishimura1

1Electrical Engineering, Stanford University, Stanford, CA, United States; 2Palo Alto Medical Foundation, Palo Alto, CA, United States

Magnetization-prepared 3D SSFP sequences have shown promise for non-contrast-enhanced flow-independent angiography (FIA), where intrinsic tissue parameters such as T1, T2, and chemical shifts are exploited to generate stable vessel contrast even under slow flow conditions. However, an important challenge with this approach is sufficient artery-vein contrast, which is crucial for artery visualization and the diagnosis of arterial disease. In this work, we apply the Maximally Stable Extremal Regions (MSER) detector and k-means clustering to perform unsupervised segmentation and removal of the femoral veins in 3D FIA datasets of the lower extremities.

Abstract PDF Presentation Video

Keywords

achieved adapted additional address agent alto angiography appear applied around arterial arteries artery audience brighter candidate centroid chemical choose clean clinicians cluster clustering clusters computing conditions confusion constraints contrast coronal correctly crucial cutting dark dataset datasets designed detected detection detector determined diagnosis dimensions disease distance electrical engineering enhanced even excite excluding exploited extended extremal extremities features femoral filtering final finally fixed flow foundation fully generate gray healthy human improve improved include incorporating independent indicating initial intensity isotropic issue legs location magnetization majority matrix maximally medical minutes morphological muscle nearby obvious overall padded particularly patient performance peripheral physicists pipeline pixel pixels pool potential predetermined prepared previous prior processing projection promise reached reaching reconstructed removal removed repeated respectively robust scanner seed segment segmentation separation shape slice slices smoothly stable starting subsequent successfully sufficient suited summary table take target terminate tracked trans uniform unstable unsupervised vein veins vessel visualization volunteer volunteers white window zero