| Title | Automated retinal layer segmentation in optical coherence tomography images with intraretinal fluid |
| Authors | Wang, Luquan Li, Xiaowen Chen, Yong Han, Dingan Wang, Mingyi Zeng, Yaguang Zhong, Junping Wang, Xuehua Ji, Yanhong Xiong, Honglian Wei, Xunbin |
| Affiliation | Foshan Univ, Sch Phys & Optoelect Engn, Foshan 528000, Guangdong, Peoples R China Foshan Univ, Sch Mech Engn & Automat, Foshan 528000, Guangdong, Peoples R China Foshan Univ, Guangdong Hong Kong Macao Intelligent Micro Nano, Foshan 528000, Guangdong, Peoples R China Guangdong Prov Key Lab Anim Mol Design & Precise, Foshan, Guangdong, Peoples R China Peking Univ, Dept Biomed Engn, Beijing 100081, Peoples R China Peking Univ, Key Lab Carcinogenesis & Translat Res, Canc Hosp & Inst, Beijing 100142, Peoples R China Shanghai Jiao Tong Univ, Sch Biomed Engn, Shanghai 200030, Peoples R China South China Normal Univ, Lab Quantum Engn & Quantum Mat, Sch Phys & Telecommun Engn, Guangzhou 510006, Guangdong, Peoples R China |
| Keywords | SD-OCT IMAGES BOUNDARIES SCANS |
| Issue Date | May-2022 |
| Publisher | JOURNAL OF INNOVATIVE OPTICAL HEALTH SCIENCES |
| Abstract | We propose a novel retinal layer segmentation method to accurately segment 10 retinal layers in optical coherence tomography (OCT) images with intraretinal fluid. The method used a fan filter to enhance the linear information pertaining to retinal boundaries in an OCT image by reducing the effect of vessel shadows and fluid regions. A random forest classifier was employed to predict the location of the boundaries. Two novel methods of boundary redirection (SR) and similarity correction (SC) were combined to carry out boundary tracking and thereby accurately locate retinal layer boundaries. Experiments were performed on healthy controls and subjects with diabetic macular edema (DME). The proposed method required an average of 415 s for healthy controls and of 482 s for subjects with DME and achieved high accuracy for both groups of subjects. The proposed method requires a shorter running time than previous methods and also provides high accuracy. Thus, the proposed method may be a better choice for small training datasets. |
| URI | http://hdl.handle.net/20.500.11897/643327 |
| ISSN | 1793-5458 |
| DOI | 10.1142/S1793545822500195 |
| Indexed | SCI(E) |
| Appears in Collections: | 工学院 北京肿瘤医院 |
