Automatic Navigation Path Planning Technology For Cardiovascular Intervention Surgery

Cardiovascular disease is one of the most common causes of death in the world.The number of cardiovascular disease patients in China has been increasing continuously by years,and this kind of disease has become the first in both morbidity and mortality among all other diseases,which seriously endangers human's health.With the development of medical technology,the minimally invasive interventional procedure performs well in curing most kinds of cardiovascular disease.Due to the complicated vasculature,a reasonable preoperative planning can greatly improve the success rate of surgery and reduce the existing risks.The path planning technology in vessels is able to generate the path from the surgical portal to the lesion,which can help doctors to plan.Therefore this article studies the path planning technology in three-dimensional blood vessels,introduces the current research of centerline extraction and stenosis detection,which are the key technologies in path planning,and proposes the corresponding algorithms and solution.In the centerline extraction part,this article presents a novel centerline extraction algorithm of tubular object based on fluid dynamics.The finite volume method is used to calculate the data of steady fluid in a three-dimensional tubular object and extract the isobaric surfaces.Then calculating center point of each isobaric surface as the control point,constructing the directed tree is according to the order of pressure levels,and optimizing the directed tree structure.Finally,fitting the control points on directed tree and obtaining the centerline by connecting all the fitting points in order.In the stenosis detection part,the stenosis of blood vessels is detected by using Bernoulli's law based on the directed tree structure,which is realized by defining the stenosis inlet,stenosis outlet and stenosis node according to the flow velocity at each control point,then the stenosis rate is calculated according to the slot of velocity at stenosis node relative to the stenosis inlet and outlet.The experimental results show that the proposed algorithms are feasible.The centerline extraction algorithm has good robustness,which can accurately extract smooth and single-pixel centerline from the complicated vessel.The stenosis detection algorithm can quickly and accurately detect the stenosis of the blood vessels and estimate the stenosis rate.On the basis of the above work,the path from surgical entrance to the lesion(stenosis)is automatically planned by backtrack process on the directed tree.Finally,an existing navigational camera perspective optimization algorithm is adopted to realize the endoscopic navigation in the blood vessel.