Detection Of Ventricular Spiral Wave Tip

Cardiovascular diseases such as arrhythmia are the most common high mortality diseases in today’s society.Cardiac electrophysiological research shows that spiral wave is one of the mechanisms leading to arrhythmia,so the research on spiral wave image is also increasing.Nowadays,there is little research on spiral wave in the field of image,so this paper studies the motion characteristics of spiral wave.In this paper,the edge of ventricular spiral wave image is detected by edge detection algorithm,and then the detected edge is detected by calculating curvature and corner detection,so as to obtain the motion trajectory of spiral wave tip,which is helpful to the clinical research of medicine in the future.In this paper,based on the classical Canny edge detection algorithm,an adaptive high and low threshold method based on twice iteration is proposed for the wave tip detection of ventricular spiral wave image,and different methods are used for wave tip detection according to the different edge features of single spiral wave image and multi spiral wave image.The main work of this paper is as follows:Firstly,in the preprocessing stage of spiral wave image,aiming at the problem of large amount of spiral wave video image data used in this paper,the fast guidance filter with higher filtering efficiency is used to replace the common guidance filter for noise processing without affecting the filtering effect,and the filtering effect and time cost of fast guidance filter are analyzed.Secondly,for the wave tip detection of single spiral wave,aiming at the problem that the traditional threshold segmentation method can not meet the high and low threshold requirements of Canny algorithm,an adaptive high and low threshold method based on twice iteration is proposed to analyze and compare the effect of edge detection.Carry out the wave tip detection based on the curvature of the edge image,and use the corrosion treatment to eliminate the discontinuities in the black edge curve.Finally,the wave tip position of the single spiral wave can be accurately detected.Finally,the single spiral wave video is detected frame by frame to obtain the wave tip trajectory of the spiral wave.Thirdly,for the wave tip detection of multi spiral wave,aiming at the problem of low overall gray level and more edge noise of multi spiral wave image,this paper proposes an adaptive high and low threshold transformation method based on secondary iteration,which is used for edge detection.The Harris algorithm is use d to detect the wave tip of the edge image,and the high and low thresholds are set to screen the corner points to obtain the wave tip position of the spiral wave.Finally,the multi spiral wave video is detected frame by frame to obtain the wave tip traje ctory of the spiral wave.Experiments show that the fast guided filtering method used in this paper significantly improves the filtering efficiency without reducing the filtering effect in the spiral wave image preprocessing stage.Compared with the tradit ional iterative threshold method,the adaptive high and low threshold method based on the second iteration obtains a more complete spiral wave edge image.The wave tip detection based on this edge image can accurately detect the wave tip position of single spiral wave and multi spiral wave,and obtain the motion trajectory of spiral wave tip.