Study Of Intraoperative Navigation In Percutaneous Coronary Interventional

At present,cardiovascular disease(CVD)is prevalence and its mortality rate rises continuously,it has become a serious threat to human health.Percutaneous coronary intervention(PCI)is a widely treatment for cardiovascular disease.This method has the advantages of less surgical trauma,shorter operative time and higher safety.Intraoperative navigation based on X-ray images is an important part of percutaneous coronary intervention.At present,it is popular to use master-slave vascular interventional surgery robot to complete the delivery of surgical instruments,which can reduce the radiation of X-ray to doctors.However,it greatly restricts the application of minimally invasive surgery robot in clinical practice because of the non-real time and low accuracy visual feedback.This paper is mainly based on the clinical demand,carry out research on X-ray image navigation operation based on the laboratory developed coronary artery interventional surgery robot based on the platform,to provide real-time accurate visual feedback for surgical robot,to effectively reduce the operation risk,improve the operation accuracy.The tracking device and control method of "surgical intervention combined with the National Natural Science Fund Project" robot assisted vessel,the main work of this paper includes: the intraoperative X angiography image navigation process,based on the X-ray images of vascular interventional device detection and contour detection algorithm research related to coronary intervention.The specific research contents and contributions include: firstly,aiming at the low accuracy,the traditional visual feedback guided wire detection in the real-time problem,this paper uses Ada Boost classifier based on LBP features to detect guide-wires,the algorithm can automatically detect the guide-wire,effectively improve the timeliness of visual feedback due to the interference,the skeletal outline of similar guide wire in the X images,the detection accuracy is low,cannot meet the clinical needs;secondly,based on machine learning guide-wire detection based on the problem of low precision,this paper uses deep learning based on the guide-wire is detected,the algorithm achieved satisfactory results in real-time and accurate detection of the guide-wire;finally,the contour detection and intraoperative navigation is another important link for the contour In this paper,a multi-scale image enhancement algorithm is used to enhance the X-ray image,and the image is detected by the classical graph cut method based on the enhanced images.Multi-scale enhancement algorithm can effectively improve the accuracy of the detection of blood vessel contour.Based on the X-ray image processing,it is easy to detect surgical instruments(guide-wire)and vessel contours to obtain the location information,which can provide real-time and accurate visual feedback for doctors.