多核多示例学习的糖尿病视网膜病变诊断  

Multi-kernel multi-instance learning based diabetic retinopathy diagnosis

作　　者：任福龙 曹鹏 杨金柱[1,2] 万超 赵大哲[1,2] Ren Fulong1,3, Cao Peng1,2, Yang Jinzhu1,2, Wan Chao4, Zhao Dazhe1,2(1. Computer Science and Engineering, Northeastern University, Shenyang 110819, China; 2. Key Laboratory of Medical Image Computing of Ministry of Education, Northeastern University, Shenyang 110819 ;3. The State Key Laboratory for New Technology of Software Architecture, Shenyang 110179, China; 4. Department of Ophthalmology, the First Hospital of China Medical University, Shenyang 110001, Chin)

机构地区：[1]东北大学计算机科学与工程学院,沈阳110819 [2]医学影像计算教育部重点实验室,沈阳110819 [3]软件架构新技术国家重点实验室,沈阳110179 [4]中国医科大学附属第一医院眼科,沈阳110001

出　　处：《中国图象图形学报》2018年第4期552-563,共12页Journal of Image and Graphics

基　　金：国家自然科学基金项目(61502091);国家高技术研究发展计划(863计划)基金项目(2015AA020106);中央高校基本科研业务费基金项目(N161604001;N150408001);沈阳市科技计划基金项目(17-134-8-00)~~

摘　　要：目的在传统糖尿病视网膜病变(糖网)诊断系统中,微动脉瘤和出血斑病灶检测的精确性决定了最终诊断性能。目前的检测诊断方法为了保证高敏感性而产生了大量假阳性样本,由于数据集没有标注病灶区域导致无法有效地建立监督性分类模型以去除假阳性。为了解决监督性学习在糖网诊断中的问题,提出一种基于多核多示例学习的糖网病变诊断方法。方法首先,检测疑似的微动脉瘤和出血斑病灶区域,并将其视为多示例学习模型中的示例,而将整幅图像视为示例包,从而将糖网诊断转化为多示例学习问题;其次,提取病灶区域的特征对示例进行描述,并通过极限学习机(ELM)分类算法过滤不相关示例以提升后续多示例学习的分类性能;最后,构建多核图的多示例学习模型对健康图像和糖网病变图像进行分类,以实现糖网病变的诊断。结果通过对国际公共数据集MESSIDOR进行糖网病变诊断评估实验,获得的准确率为90.1%,敏感性为92.4%,特异性为91.4%,ROC(receiver operating characteristic)曲线下面积为0.932,相比其他算法具有较大性能优势。结论基于多核多示例学习方法在无需提供病灶标注的情况下,能够高效自动地对糖网病变进行诊断,从而既能避免医学图像中标注病灶的费时费力,又可以免除分类算法中假阳性去除的问题,获得较好的效果。Objective Diabetic retinopathy （DR） is one of the complications of diabetes and causes severe vision loss and blindness in severe cases if left untreated. A regular eye examination is important for initial diagnosis and early treatment. The change in the blood vessels of the retina is the leading cause of DR. The form of red Lesions, such as hemorrhage/microaneurysm （HMA）, is the first explicit sign and an important symptom of DR. Hence, in the traditional DR diagnosis system, the accuracy of HMA lesion detection determines the final diagnosis performance. The diagnosis method produces a large number of false positive samples for high sensitivity, and the supervised classification model is ineffective in removing false positives because the dataset does not label the lesion area. A new algorithm based on multi-kernel and multi-instance learning is proposed to solve the problem of supervised learning in DR diagnosis. Method First, a multi-scale morphological top-hat transformation is employed to enhance blood vessels and red lesions on the green channel image, then the main vessels of the retina are segmented by thresholding technique on the mask image obtained by binarizing the enhanced image. All regions of interest are generated by subtracting the main vessels from the mask image, and a connected-component labeling technique based on region growing is conducted to detect the suspicious HMA. The detected HMA areas are considered instances, and the entire image is considered a bag. Thus, the problem of DR diagnosis is considered a multi-instance learning problem. Second, a 37D feature based on color, texture, and shape is extracted for each candidate HMA to describe the instance in multi-instance learning. Numerous suspected HMAs are generally obtained to ensure high sensitivity in the initial detection of the lesions, but many HMAs would produce a negative effect on the performance of the multi-instance learning. An extreme learning machine （ELM） -based classifier is accordingly constructed to filter

关 键 词：糖尿病视网膜病变 微动脉瘤 眼底图像 计算机辅助诊断 多示例学习 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]