Three-dimensional Reconstruction Of Coronary Artery And Assessment Of Stenosis

Coronary heart disease is one of the main threats to human health.In China,the mortality rate of coronary heart disease has been on the rise.At present,many methods have been developed in the medical community to diagnose coronary heart disease.Electrocardiogram,echocardiography,CTA,etc.can indirectly reflect whether coronary has disease or not.In 1993,Nico Pijls et al.Proposed a new index for estimating coronary blood flow through pressure measurement-Fractional Flow Reserve(FFR).After long-term basic and clinical research,FFR has become a recognized indicator of functional assessment of coronary stenosis.However,traditional FFR requires the insertion of a catheter in the coronary artery to calculate the pressure in the stenosis,which is an invasive diagnosis and has very limited limitations.In recent years,research on FFR-CT is more and more popular.This method uses coronary computed tomography angiography(CCTA)images to perform three-dimensional reconstruction of coronary arteries,and uses computational fluid dynamics(CFD)for finite element numerical simulation to calculate bleeding Flow Reserve Fraction.This method is non-invasive and has wide applicability.This paper based on the CTA images of coronary arteries,researches a three-dimensional reconstruction algorithm of coronary arteries and the finite element analysis which is used to calculate the blood flow reserve fraction to evaluate the degree of coronary stenosis.This paper uses a Hessian-based filter to filter the CTA image,which significantly enhances the tubular structure,and obtains the filtered vessel enhanced image and the direction information of the vessel.The shortest path-based method is used to extract the centerline of the blood vessel,and the direction information of the voxels in the blood vessel is creatively used to make the centerline more accurate.Based on the traditional graph cut algorithm,we modify the edge energy term based on the difference in the intensity of adjacent voxels,and the inner product of the blood vessel directions of two adjacent voxels is also multiplied.Compared with the traditional graph-cutting algorithm,our anisotropic graph-cutting algorithm effectively reduces the unnatural bulges on the vascular cavity,resulting in a more accurate coronary model.The above algorithm only needs to manually select two seed points,and the others are automatically completed by the computer,which is a semi-automatic blood vessel segmentation algorithm.Finally,we use the software to smoothly process the coronary artery model and mesh the model.The finite element analysis method is used to calculate the blood flow reserve fraction(FFR)to evaluate the degree of coronary artery stenosis.