[1]高付林,倪森,付冬梅.运用像素值分析测量法计算眼底出血面积[J].眼科新进展,2017,37(1):046-48.[doi:10.13389/j.cnki.rao.2017.0012]
　GAO Fu-Lin,NI Sen,FU Dong-Mei.Using pixel value measurement method to calculate area of fundus hemorrhage[J].Recent Advances in Ophthalmology,2017,37(1):046-48.[doi:10.13389/j.cnki.rao.2017.0012]
点击复制

运用像素值分析测量法计算眼底出血面积/HTML
分享到：

《眼科新进展》[ISSN:1003-5141/CN:41-1105/R]

卷:
37卷
期数:
2017年1期
页码:
046-48
栏目:
应用研究
出版日期:
2017-01-05

文章信息/Info

Title:
Using pixel value measurement method to calculate area of fundus hemorrhage
作者:
高付林; 倪森; 付冬梅
100101　北京市，解放军306医院眼科（高付林）；100083　北京市，北京科技大学控制科学与工程系（倪森，付冬梅）
Author(s):
GAO Fu-Lin; NI Sen; FU Dong-Mei
Department of Ophthalmology，the 306th Hospital of PLA(GAO Fu-Lin)，Beijing 100101，China;the School of Automation and Electrical Engineering，University of Science and Technology Beijing(NI Sen，FU Dong-Mei)，Beijing 100083，China
关键词:
像素值; 分析测量法; 眼底出血面积
Keywords:
image pixels; pixel value measurement; retinal hemorrhage area
分类号:
R770.4
DOI:
10.13389/j.cnki.rao.2017.0012
文献标志码:
A
摘要:
目的　运用像素值分析测量法自动检测眼底出血面积，量化眼底出血评价标准。方法　应用不同彩色通道特点选择样本库中眼底出血的图像28幅，根据视盘、黄斑、血管的颜色和出血形状特性选取绿色通道（G通道）设定阈值来提取出血，Matlab软件进行图像分析，用office表格软件进行数据统计分析，计算出血区域占眼底图像面积的百分比，并计算其准确率，评价算法的精度。结果　6幅眼底出血图像的出血面积百分比分别为6.26%、4.21%、10.74%、2.01%、7.73%、5.01%，相对误差平均值3.61%。结论　运用像素值分析测量法自动检测眼底出血面积的方法能够相对完整和准确地提取眼底图像出血目标，且时间效率高，保证了算法精度。
Abstract:
Objective　To automatically detect the area of retinal hemorrhage by using the pixel value measurement method,and quantify the retinal hemorrhage criteria.Methods　Twenty-eight retinal hemorrhage images were chosen to extract the bleeding characteristics.The G color channel in fundus images by statistical properties of the retinal disc,macular,vascular,hemorrhage was used to set threshold extract the hemorrhage part.Matlab software was used to analysis image features.Office software was used to analyze pixels data.Finally,the percentage of fundus hemorrhage area was calculated.Meanwhile,the accuracy of the algorithm was evaluated.Results　The percentage of fundus hemorrhage area in 6 images were 6.26%,4.21%,10.74%,2.01%,7.73% and 5.01%,respectively,the average relative error was 3.61%.Conclusion　Using the pixel value measurement method can relatively extract the retinal hemorrhage targets completely and accurately,and can ensure the algorithm accuracy with high computational efficiency.

参考文献/References:

［1］　殷晓贝,李根林.视网膜色素变性的眼底血流变化研究现状［J］.眼科新进展,2015,35（5）:482-484,488.
YIN XB,LI GL.Research advances in fundus blood flow changes in retinitis pigmentosa［J］.Rec Adv Ophthalmol,2015,35(5):482-484,488.
［2］　TANG L，NIMEIJER M，REINHARDT JM,GARVIN MK,ABRM-OFF MD.Splat feature classification with application to retinal hemorrhage detection in fundus images［J］.IEEE Trans Med Imaging,2013,32(2):364-375.
［3］　王红星,韩倩,张琰,于广委,李筱荣.眼底像血管管径测量与眼病［J］.眼科新进展,2014,34（2）:197-200.
WANG HX,HAN Q,ZHANG Y,YU GW,LI XR.Fundus measurement like vessel diameter and eye disease［J］.Rec Adv Ophthalmol,2014,34(2):197-200.
［4］　付校楠,彭惠.眼底自发荧光技术在部分视网膜疾病中的应用进展［J］.眼科新进展,2013,33(4):393-396.
FU XN,PENG H.Application of fundus autofluorescence technique in some retinal diseases［J］.Rec Adv Ophthalmol,2013,33(4):393-396.
［5］　荣华,周晓东.丹参酮ⅡA磺酸钠治疗眼底出血48例［J］.眼科新进展,2008,28(9):702-703.
RONG H,ZHOU XD.Treatment of fundus hemorrhage with tanshinone ⅡA sodium sulfonate in 48 cases［J］.Rec Adv Ophthalmol,2008,28(9):702-703.
［6］　QURESHI RJ,KOVACS L,HARANGI B,NAGY B,PETO T,HAJDU A.Combining algorithms for automatic detection of optic disc and macula in fundus images［J］.CVIU,2012,116(1):138-145.
［7］　MICHAEL A,GINNEKEN BV,MEINDERT N,ABRAMOFF MD,BRAM VG.Automatic detection of red lesions in digital color fundus photographs:U.S.Patent 7,474,775［P］.2009-1-6.
［8］　郑绍华,陈健,潘林,郭健,余轮.眼底图像中黄斑中心与视盘自动检测新方法［J］.电子与信息学报,2014,11:2586-2592.
ZHENG SH,CHEN J,PAN L,GUO J,YU L.A novel method of macula fovea and optic disk automatic setection for retinal images［J］.JEIT,2014,36(11):2586-2592.
［9］　彭英辉,张东波,沈奔.基于多尺度匹配滤波和集成学习的眼底图像微脉瘤检测［J］.计算机应用,2013,2:543-546，566.
PENG YH,ZHANG DB,SHEN B.Microaneurysm detection based on multi-scale match filtering and ensemble learning［J］.J Comput Appl,2013,2:543-546,566.
［10］　肖志涛,赵北方,张芳,耿磊,吴骏,王雯,等.基于k均值聚类和自适应模板匹配的眼底出血点检测方法［J］.中国生物医学工程学报,2015,34(3):264-271.
XIAO ZT,ZHAO BF,ZHANG F,GENG L,WU J,WANG W，et al.Hemorrhages detection in fundus image based on k-means clustering and adaptive template matching［J］.CJBE,2015,34(3):264-271.

备注/Memo

备注/Memo:
地方首都特色面上项目 （编号：Z141107002514056）
更新日期/Last Update: 2017-02-15