Extracting contrast-filled vessels in X-ray angiography by graduated RPCA with motion coherency constraint

详细信息    查看全文

• 作者：Mingxin Jin^a ; ¹ ; Rong Li^b ; ¹Author Vitae ; Jian Jiang^b ; Binjie Qin^a ; ^{bjqin@sjtu.edu.cn}Author Vitae
• 关键词：Subspace estimation ; Robust principal component analysis (RPCA) ; Low-rank model ; Spatio-temporal motion coherency ; Matrix decomposition ; Trajectory decomposition ; X-ray coronary angiograms
• 刊名：Pattern Recognition
• 出版年：2017
• 出版时间：March 2017
• 年：2017
• 卷：63
• 期：Complete
• 页码：653-666
• 全文大小：2692 K
• 卷排序：63

文摘

We propose a graduated RPCA with motion coherency constraint to extract contrast-filled vessels from X-ray coronary angiography. A statistically structured RPCA exploits the complex structural connectivity to identify all candidate vessel. Total variation & L1-based regularized RPCA is implemented on candidate vessels for accurate vessel detection. Qualitative and quantitative evaluations demonstrate the obvious advantages of our method over the state-of-the-art methods.