Mutual Structure Model for improving vessel quality in X-ray angiography

To facilitate the analysis and diagnosis of X-ray angiography in interventional surgery, some efforts have attempted to hessian the X-ray vessel. However, vessel images of angiography suffered from low quality and presence of large artifacts prevents the clinical application of technology. In this paper, we propose an effective vessel framework that improve original vessel and hessian vessel quality in angiographic data. In this framework, we propose mutual structure model for increasing the vessel quality. Through experiments conducted on angiographic data, we demonstrated the excellent performance of the proposed method in removing motion artifacts and background structures. We evaluate our method for the different vessel structure from clinical scene, and show competitive results as compared with two hessian techniques in terms of qualitative and quantitative evaluation.