基于自动化三维超声冠状面图像的腹壁手术切口检测算法

Algorithm for abdominal surgery incision detection based on automated 3D ultrasound coronal plane images

切口疝是发生于原手术切口区域的腹壁缺损,手术切口的准确定位对于切口疝的诊断具有重要临床意义。本文提出一种基于腹壁三维超声冠状面图像的手术切口检测算法,可辅助医生快速定位切口疝病灶区域。算法首先采用斑点降噪各向异性扩散算法对冠状面超声图像进行滤波处理;然后再进行图像预处理操作以减少算法运算量,提高算法执行效率;最后通过边缘检测、最大距离筛选和线性拟合操作获得原手术切口的直线方程。结果表明:算法拟合的手术切口直线与实际手术切口位置非常接近,拟合效果好,精确度高。

Incisional hernia is an abdominal wall defect that occurs in the original incision area.Accurate positioning of surgery incision has important clinical value for incisional hernia diagnosis.In this paper,an algorithm is proposed for abdominal surgery incision detection based on automated 3D ultrasound coronal plane images.Firstly,the coronal plane ultrasound images are filtered by using Speckle Reducing Anisotropic Diffusion method.Then,the image pre-processing operations are used to reduce the amount of calculation and improve the execution efficiency.Finally,a straight-line equation of the original surgery incision is obtained by using edge detection,maximum distance screening and linear fitting.It is shown in the results that the fitting line of the surgery incision is very close to the location of the actual surgery incision,which means the algorithm has satisfied fitting features and high detection precision.