无线电工程2025，Vol.55Issue(2)：424-435,12.DOI:10.3969/j.issn.1003-3106.2025.02.023
基于骨干检测算法的脑部核磁影像辅助诊断
Brain Magnetic Resonance Imaging Assisted Diagnosis Based on Backbone Detection Algorithm
摘要
Abstract
With the development of medical imaging technology,the auxiliary diagnosis system plays an important role in improving the efficiency of doctors.In order to better assist doctors in the diagnosis of brain diseases,the development status of medical image segmentation is briefly introduced.Based on Canny algorithm and Otsu algorithm,a novel graph backbone structure detection algorithm is proposed by introducing clustering idea.The algorithm completes the edge detection task efficiently by setting the gray threshold,determining the origin of the image,performing multi-directional scanning,pixel-by-pixel comparison,and outputting the backbone structure.The eight-direction backbone structure detection experiment is carried out on the brain magnetic resonance images of different patients.The results show that compared with the existing algorithms of Roberts,Prewitt,Sobel,Canny and Otsu,the backbone detection algorithm reduces the processing time to 0.70 s by changing the time complexity to O(w+h)under the premise of ensuring the accuracy of the results.Because the proposed algorithm does not need to collect a large number of data samples like deep learning methods,and does not consume server computing power,it greatly reduces the processing cost while improving the efficiency of computer-aided diagnosis.In the future,it will show great application potential in primary hospitals.关键词
脑部核磁影像/骨干检测算法/聚类分割思想/边缘检测Key words
brain magnetic resonance imaging/backbone detection algorithm/clustering segmentation idea/edge detection分类
信息技术与安全科学引用本文复制引用
白世强,陈楷,李丽宏..基于骨干检测算法的脑部核磁影像辅助诊断[J].无线电工程,2025,55(2):424-435,12.基金项目
河北省自然科学基金(F2023402011) （F2023402011）
邯郸市科学技术局项目(21422031251)Hebei Provincial Natural Science Foundation of China(F2023402011) （21422031251）
Handan Science and Technology Bureau Project(21422031251) （21422031251）
