Algorithm Research Of 3D Vessel Segmentation In CTA Image Of Kidney

In recent years,the increase of renal cancer make renal tumor resection surgery become the preferred alternative of local renal cancer.To ensure the live quality of patients after surgery,partial nephrectomy surgery of segmental renal artery clamping has been used into treatment increasingly.In the implementation of this surgery,we require to block the renal aorta and clearly show the renal artery branch to choose a reasonable block arterial branch of surgery.Because of the data volume of kidney CTA(Computed Tomographic Angiography)image is large,manual division brings a lot of work to doctor,from the perspective of improving the efficiency of work and imaging accuracy,this thesis proposed a method to automate enhance renal artery and segment the aortaventralis and renal aorta.In order to accurately and clearly show the renal artery,especially those smaller renal artery branch distal whice close to or penetrate into the kidney tumor,this thesis presents a renal artery enhancement method based on symmetry matched filter,in this method,we design a set of multi scale and orientation symmetry matched filter conform to 3D vessel’s characteristic,enhance renal artery and suppress the wrong response of renal cortex.In this thesis,an automatic segmentation method based on Hough transform and regional growth is proposed for the extraction of arterial artery between aortaventralis and aortaventralis and kidney.First,the CTA image is displayed in two directions,the gray value of the aortaventralis is high and its shape is similar to the circular shape.The Hough transform is used to calculate the center and radius of the aorta of each frame of the 3D CTA image.Then,using the segmentation of the kidney area as a mask,a random algorithm was used to obtain the seed points in the kidney area and set appropriate growth threshold to extract renal aorta between aortaventralis and kidney,and then complete the automatic extraction solution.This thesis use synthetic data and clinic CTA data for experimental verification and quantitative analysis to present applicability and effectiveness of the method we proposed and improve the work efficiency of doctors using this method as automatic algorithm of computer-aided diagnosis.