融合特征相互关系的视网膜微动脉瘤提取  

Retinal microaneurysm extraction by fusing relationship among features

作　　者：赖小波[1] 刘华山[2] 方纯洁[1] 

机构地区：[1]浙江中医药大学信息技术学院,浙江杭州310053 [2]东华大学信息科学与技术学院,上海201620

出　　处：《光学精密工程》2013年第8期2187-2194,共8页Optics and Precision Engineering

基　　金：国家自然科学基金资助项目(No.61203337);浙江省自然科学基金资助项目(No.LQ12F01004);浙江中医药大学校级科研基金资助项目(No.2012ZY18)

摘　　要：为了抑制视网膜不同结构特征之间的影响,提高视网膜微动脉瘤的检测精度,提出了一种基于特征相互关系的视网膜微动脉瘤提取算法。首先,对视网膜灰度图像进行均值滤波,检测圆形边界和视盘,并构建视盘掩模。然后,对视网膜绿色分量图像自适应直方图均衡化,利用Canny方法提取边缘,移除图像圆形边界并填充封闭的小面积对象。最后,考虑不同特征之间的相互关系,消除较大面积对象后进行"逻辑与"运算移除视网膜渗出物、血管和视盘,得到视网膜微动脉瘤图像。实验结果表明:该算法能够有效提取视网膜眼底图像中的微动脉瘤,其敏感度、特异性、阳性预测值和检测精度分别达到了94.81%、96.04%、91.64%和95.66%,基本能够满足临床应用对稳定性和精度的要求。To suppress the mutual affects among different structure features of retinal and improve the detection precision of retinal microaneurysms, a microaneurysm extraction algorithm by fusing rela- tionship among features was proposed. Firstly, the mean filter was applied to a retinal grayscale im- age, both the circular border and optic disc were detected, and the optic disc mask was created. Then, the green component of the retinal image was equalized with an adaptive histogram and Canny method was used to extract the edges before removing the image circular border and to fill the enclosed small area objects. Finally, with consideration of the relationship among different features, larger area ob- jects were removed and an ＇AND＇ logic was used to remove the retinal exudates, blood vessels as well as optic disc to obtain the retinal microaneurysm image. Experimental results indicate that the pro- posed method can effectively extract the microaneurysms in the retinal fundus image, and their sensi- tivity, specificity, positive predictive value and accuracy are 94. 81%, 96. 04%, 91. 64 % and 95.66~, respectively. It can satisfy the clinical application requirements for strong stabilization and higher precision.

关 键 词：视网膜图像 微动脉瘤 特征提取 特征相互关系 自适应直方图均衡化 

分 类 号：R774.1[医药卫生—眼科] TP391.4[医药卫生—临床医学]