论文部分内容阅读
目的计算和测量上肢静脉超声图像中穿刺靶血管的深度和管径信息,为实现自动穿刺奠定基础。方法采用两种静脉横截面超声图像中管径的测量方法,一是基于评估后自适应选取的阈值分割血管,然后上下分别扫描静脉血管上下边缘,得到管径信息。二是基于大津(Otsu)法分割血管,运用形态学运算对二值图像进行预处理,然后用与方法一类似的方式计算管径信息,结合管径信息再对静脉的深度信息进行矫正。结果在500张上肢静脉超声图像上测试上述两种方法,方法一、二的准确率分别为93.0%和99.4%。结论基于大津和形态学处理的方法对上肢静脉超声图像管径测量效果更佳。
Objective To calculate and measure the depth and diameter information of the target vessel in the superior vena cava ultrasound images and lay the foundation for automatic puncture. Methods Two methods of measuring the diameter of the vein in the cross-sectional ultrasound images were used. One was to divide the blood vessels based on the thresholds adaptively selected after the assessment, and then scan the upper and lower edges of the veins respectively to get the diameter information. The second is to divide the blood vessels based on the Otsu method. Morphological operation is used to preprocess the binary images. Then the diameter information is calculated in a way similar to the method one, and the vein depth information is corrected with the diameter information. Results The above two methods were tested in 500 images of upper extremity venous ultrasound. The accuracy of method one and two were 93.0% and 99.4% respectively. Conclusion The method of Otsu and morphological treatment is more effective in measuring the diameter of the upper extremity venous ultrasound image.