Diagnosis Of Coronary Heart Disease By Optical Coherence Tomography Using Random Walk Algorithm

Objective: The objective is to study the diagnosis of coronary heart disease by optical coherence tomography based on random walk algorithm, so as to assist doctors in diagnosing coronary heart disease. Method: K-means algorithm combined with mathematical morphology provides seed points for random walk algorithm, and realizes semi-automatic segmentation of different kinds of patches. Secondly, by adding the space distance term based on the distance between the edge and the seed point in the weight function, the random walk algorithm is improved, which can make the weak edge patch area not be segmented too much. Results: the results were obtained by comparing the traditional random walk method and the literature method. The segmentation accuracy of this method is higher, and it is more convenient to use. Moreover, the average segmentation time of single image is not more than 5 s, which can basically meet the real-time needs of clinical diagnosis. Conclusion: Experiments show that the method proposed in this study is easy to use, with high segmentation accuracy, and to some extent, it can provide the clinician with plaque area information, improve the speed of reading films, and assist doctors in diagnosing coronary heart disease. In this study, the random walk algorithm is improved to get a better segmentation algorithm than other methods.