北京交通大学学报2017，Vol.41Issue(6)：13-20,8.DOI:10.11860/j.issn.1673-0291.2017.06.003
基于多尺度区域块的糖尿病性视网膜病变级联检测
Cascade detection of diabetic retinopathy based on multi-scale region blocks
摘要
Abstract
At present,morphological methods for detecting diabetic retinopathy have high com-plexity.The traditional deep learning methods avoid the exclusion of physiological structure, manual design features.However,they have to do large calculation and the speed is relatively slow.In order to solve these problems,this paper presents a cascade detection framework based on deep learning.Firstly,the fundus images are divided into blocks to detect whether there are lesions,and then pixels in these lesions are classified into four categories:microaneurysms, haemorrhages,hard exudates and soft exudates.The experimental results show that on the public DIARETDB1 fundus image database,the detection of four kinds lesions with sensitivity are 88.62％,94.91％,98.91％ and 92.91％,respectively.Compared to morphological methods,the accuracy has improved 17.39％ in microaneurysms and 15.18％ in haemorrhages.And the detec-tion time is only a quarter of the traditional deep learning methods.关键词
信号与信息处理/糖尿病性视网膜病变/卷积神经网络/眼底图像/病变分割Key words
signal and information processing/diabetic retinopathy/convolution neural network/fundus image/lesion segmentation分类
信息技术与安全科学引用本文复制引用
马文婷,赵耀,韦世奎,张诗吟,廖理心..基于多尺度区域块的糖尿病性视网膜病变级联检测[J].北京交通大学学报,2017,41(6):13-20,8.基金项目
国家自然科学基金(61532005,61572065) （61532005,61572065）
国家重点研发计划(2016YFB0800404) （2016YFB0800404）
教育部-中国移动联合基金(MCM20160102) Foundation items:National Nature Science Foundation of China(61532005,61572065 ) （MCM20160102）
National Key Research and Development Plan of China(2016YFB0800404) （2016YFB0800404）
Joint Fund of Ministry of Education of China and China Mobile(MCM20160102) （MCM20160102）
