Automated Detection of Malarial Retinopathy in Digital Fundus Images for Improved Diagnosis in Malawian Children with Clinically Defined Cerebral Malaria.
Automated Detection of Malarial Retinopathy in Digital Fundus Images for Improved Diagnosis in Malawian Children with Clinically Defined Cerebral Malaria.
复制标题
DOI：
10.1038/srep42703
复制
发表时间：
2017-02-15
影响因子：
4.6
通讯作者：
Harding SP
中科院分区：
文献类型：
--
作者：
Joshi V;Agurto C;Barriga S;Nemeth S;Soliz P;MacCormick IJ;Lewallen S;Taylor TE;Harding SP
脑型疟疾（CM）是疟疾感染的一种并发症，是非洲儿童大多数疟疾相关死亡的原因。 CM 的标准临床病例定义错误分类了约 25% 的患者，但当临床病例定义中添加疟疾视网膜病变 (MR) 时，特异性从 61% 提高到 95%。眼底镜检查需要昂贵的设备和技术专业知识，这在疟疾流行地区通常不具备，因此我们开发了一种自动化软件系统来分析视网膜彩色图像的 MR 病变：视网膜变白、血管变色和白色中心出血。使用偏最小二乘分类器组合各个病变检测算法来确定 MR 的存在或不存在。我们使用了 86 名临床定义的 CM 儿科患者（70 名患有 MR，16 名没有 MR）的回顾性视网膜图像数据集来评估算法性能。我们的目标是降低 CM 诊断的假阳性率，因此对算法进行了高特异性调整。整体 MR 检测的灵敏度/特异性为 95%/100%，视网膜变白的灵敏度/特异性为 65%/94%，血管变色的灵敏度/特异性为 62%/100%，出血的灵敏度/特异性为 73%/96%。这种使用视网膜彩色图像检测 MR 的自动化系统有可能提高 CM 诊断的准确性。
Cerebral malaria (CM), a complication of malaria infection, is the cause of the majority of malaria-associated deaths in African children. The standard clinical case definition for CM misclassifies ~25% of patients, but when malarial retinopathy (MR) is added to the clinical case definition, the specificity improves from 61% to 95%. Ocular fundoscopy requires expensive equipment and technical expertise not often available in malaria endemic settings, so we developed an automated software system to analyze retinal color images for MR lesions: retinal whitening, vessel discoloration, and white-centered hemorrhages. The individual lesion detection algorithms were combined using a partial least square classifier to determine the presence or absence of MR. We used a retrospective retinal image dataset of 86 pediatric patients with clinically defined CM (70 with MR and 16 without) to evaluate the algorithm performance. Our goal was to reduce the false positive rate of CM diagnosis, and so the algorithms were tuned at high specificity. This yielded sensitivity/specificity of 95%/100% for the detection of MR overall, and 65%/94% for retinal whitening, 62%/100% for vessel discoloration, and 73%/96% for hemorrhages. This automated system for detecting MR using retinal color images has the potential to improve the accuracy of CM diagnosis.