摘要
ObjectiveDiabetic retinopathy(DR)is a serious complication of diabetes that may cause vision loss or evenblindness in patients.Early examination of the choroid plays an essential role in the diagnosis of DR.However,owing to the fuzzychoroid-sclera interface(CSI)and shadow of retinopathy in the optical coherence tomography(OCT)images of DR,most existingalgorithms cannot segment the choroid layer precisely.The present paper aims to improve the accuracy of choroid segmentation inDR OCT images.MethodsIn this paper,we propose an optimized squeeze-excitation-connection(SEC)module integrated withthe UNet,called the SEC-UNet,which not only focuses on the target but also jumps out of the local optimum to enhance the overallexpressive ability.ResultsThe experimental results show that the area under the ROC curve(AUC)of the SEC-UNet reaches up to0.9930,which outperforms that obtained for conventional UNet and SE-UNet models.It indicates that the SEC-UNet can obtainaccurate and complete segmentation results of the choroid layer.Statistical analysis of choroid parameter changes indicated thatcompared with normal eyes,the1mm adjacent area of choroid fovea increased in87.1% of DR patients.It proved that DR is likelyto cause choroid layer thickening.ConclusionOur method may become a useful diagnostic tool for doctors to explore the functionof the choroid in the prevention,pathogenesis,and prognosis of diabetic eye disease
学科与关键词
- Choroidal vascularity index (CVI)-Net-based automatic assessment of diabetic retinopathy severity using CVI in optical coherence tomography imagesWang, Xuehua; Li, Rui; Chen, Junyan; Han, Dingan; Wang, Mingyi; Xiong, Honglian; Ding, Wenzheng; Zheng, Yixu; Xiong, Ke*; Zeng, Yaguang*Journal of Biophotonics, 2023, 16(6).
- Automatic Segmentation and Measurement of Choroid Layer in High Myopia for OCT Imaging Using Deep LearningXu, Xiangcong; Wang, Xuehua*; Lin, Jingyi; Xiong, Honglian; Wang, Mingyi; Tan, Haishu; Xiong, Ke*; Han, Dingan*Journal of Digital Imaging, 2022, 35(5): 1153-1163.被引 2
- EAD-Net: A Novel Lesion Segmentation Method in Diabetic Retinopathy Using Neural NetworksWan, Cheng; Chen, Yingsi; Li, Han; Zheng, Bo; Chen, Nan; Yang, Weihua*; Wang, Chenghu*; Li, Yan*DISEASE MARKERS, 2021, 2021: 6482665.被引 33
