期刊
IET IMAGE PROCESSING
卷 14, 期 6, 页码 1081-1090出版社
INST ENGINEERING TECHNOLOGY-IET
DOI: 10.1049/iet-ipr.2019.1007
关键词
medical image processing; biomedical optical imaging; image segmentation; blood vessels; learning (artificial intelligence); eye; sensitivity analysis; retinal blood vessels; diagnostic biomarker; unified loss function; patch-based generative adversarial network-based technique; generator network; conditional patch-based generative adversarial network; retinal vessel segmentation; CPGAN; deep learning methods; diabetic retinopathy; ophthalmologic retinopathy; spatial features; biased distribution; fundoscopic images; receiver operating characteristic curves
类别
资金
- key project of the National Natural Science Foundation of China [91630206]
Retinal blood vessels, the diagnostic bio-marker of ophthalmologic and diabetic retinopathy, utilise thick and thin vessels for diagnostic and monitoring purposes. The existing deep learning methods attempt to segment the retinal vessels using a unified loss function. However, a difference in spatial features of thick and thin vessels and a biased distribution creates an imbalanced thickness, rendering the unified loss function to be useful only for thick vessels. To address this challenge, a patch-based generative adversarial network-based technique is proposed which iteratively learns both thick and thin vessels in fundoscopic images. It introduces an additional loss function that allows the generator network to learn thin and thick vessels, while the discriminator network assists in segmenting out both vessels as a combined objective function. Compared with state-of-the-art techniques, the proposed model demonstrates the enhanced accuracy, sensitivity, specificity, and area under the receiver operating characteristic curves on STARE, DRIVE, and CHASEDB1 datasets.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据