基于相位一致性模型的眼底微动脉瘤检测方法  

Microaneurysms Detection in Fundus Image Based on Phase Congruency

作　　者：肖志涛[1] 刘璐[1] 耿磊[1] 张芳[1] 吴骏[1] 张欣鹏[1] 苏龙[2] 陈莉明[3] 单春燕[3] 

机构地区：[1] 天津工业大学电子与信息工程学院,天津300387 [2] 天津医科大学眼科医院,天津300384 [3] 天津医科大学代谢病医院 天津300070

出　　处：《中国生物医学工程学报》2014年第1期8-15,共8页Chinese Journal of Biomedical Engineering

基　　金：国家自然科学基金(61102150);天津市科技支撑计划重点项目(13ZCZDGX02100)

摘　　要：眼底微动脉瘤是糖尿病视网膜病变最早期的症状,准确检测眼底图像中的微动脉瘤对糖尿病视网膜病变的筛查具有重要意义。提出一种基于相位一致性模型的微动脉瘤检测方法。首先采用相位一致性模型获取微动脉瘤候选者,然后通过构建灰度剖面图去除图像中血管片段等无关信息,从而筛选出真正的微动脉瘤。通过对ROC网站提供的50幅眼底图像进行实验,在图像水平上实现了灵敏度94%、特异性100%、准确率96%的检测效果。结果表明,该方法对图像的亮度、对比度不敏感,能够高效自动地检测出彩色眼底图像中的微动脉瘤。The presence of microaneurysms in the retina is the earliest clinical symptom of diabetic retinopathy （DR） , thus their reliable detection is essential in the DR screening system. Based on phase congruency, this paper proposes a new microaneurysms detection method. The first step aimed at obtaining microaneurysms candidate regions achieved by using phase congruency. Then the irrelevant information, such as the vessel fragments, was removed by constructing directional cross-section profiles. Through testing on 50 fundus images provided by ROC website, the method achieved a sensitivity of 94% , specificity of 100% , and accuracy of 96% at the image level, respectively. This method can accurately get microaneurysms in color fundus images, and it is insensitive to image brightness and contrast.

关 键 词：微动脉瘤检测 相位一致性 灰度剖面图 眼底图像 糖尿病视网膜病变 

分 类 号：R318[医药卫生—生物医学工程] TP391.4[医药卫生—基础医学]