Aortic Segmentation And Hemodynamic Simulation Analysis Based On 3D-SLIC And Shape Features

The aorta is the most important blood vessel in the human body,and cardiovascular disease mortality due to aortic dissection is high.During diagnosis and treatment,the attending physician needs to obtain accurate three-dimensional models of the external aorta and internal hemodynamic information.For the CT image sequence in the diagnosis and treatment of aortic dissection,how to accurately extract the aortic region from it is an urgent problem.This paper designs a three-dimensional segmentation method based on the combination of superpixels and shape features.By efficiently and accurately segmenting the aortic region and realizing 3D visualization,it provides effective information for doctors to perform diagnosis and treatment.Dynamic simulation analysis provides effective reference information for the pathogenesis of aortic dissection.The main work of this article is as follows:(1)The aortic segmentation data set of CT images of human thoracic and abdominal cavity was collected and established,and the aortic region was determined by manual annotation.This data set is of great significance for guiding the research of segmentation algorithm and evaluation of algorithm performance.At the same time,an automatic localization algorithm based on shape features was designed to realize the precise positioning of the aortic spur area.(2)The model was extracted based on the shape features of the aorta and the simple linear iterative clustering algorithm(SLIC)was improved.A 3D-SLIC algorithm suitable for aortic three-dimensional segmentation was designed,and the superpixel results were merged using a region adjacency matrix.Comparing the experimental results with the gold standard,the algorithm designed in this paper can accurately and efficiently segment and extract the aortic region from the three-dimensional space.(3)A three-dimensional model of the aortic region after segmentation and extraction is established,and the blood environment in the aorta is simulated through hemodynamic simulation analysis.The effects of blood flow on aorta and aortic dissection at different inlet pressures were analyzed.The pathogenesis and diagnosis and treatment of aortic dissection provide reference data.The aortic three-dimensional segmentation method designed in this paper fully considers the shape and space characteristics of the aorta,and can efficiently and accurately perform three-dimensional segmentation of the aorta.The overlap rate and time are used to quantify the results of the method in this paper.Compared with other aortic segmentation algorithms,the overlap rate is up to 10.768%,and the time consumption is reduced by nearly 3/4.Hemodynamic analysis was performed on the aortic model to simulate the blood environment in the aorta,and the influence of blood flow on the aorta and aortic dissection was analyzed to provide doctors with auxiliary support information.The work in this article is a good aid for the diagnosis and treatment of aortic dissection,it can improve the work efficiency of doctors,and it is of great significance to save patients' lives.