Self-assessing image-based respiratory motion compensation for fluoroscopic coronary roadmapping

We present a self-assessing image-based motion compensation method for coronary roadmapping in fluoroscopic images. Extending our previous work on respiratory motion compensation, we introduce kernel-based nonparametric data analysis in this work to better characterize the objective function involved in motion estimation, which leads to two new improvements in motion compensation. First, through mode analysis we are able to capture the dominant component of the respiratory image motion and increase the chance of finding the global optimum. Second, an information theoretic measure is proposed to assess the uncertainty of the motion estimation and automatically detect unreliable motion estimates. The benefits of the proposed method are shown through evaluations performed on real clinical data from different procedures of percutaneous coronary interventions.