Deep learning-based inpainting of saturation artifacts in optical coherence tomography images

作     者：Muyun Hu Zhuoqun Yuan Di Yang Jingzhu Zhao Yanmei Liang 

作者机构：Institute of Modern OpticsNankai UniversityTianjin Key Laboratory of Micro-scale Optical Information Science and TechnologyTianjin 300350China Department of Thyroid and Neck TumorTianjin Medical University Cancer Institute and Hospital National Clinical Research Center for CancerKey Laboratory of Cancer Prevention and TherapyTianjin 300060China 

出 版 物：《Journal of Innovative Optical Health Sciences》 (创新光学健康科学杂志（英文）)

年 卷 期：2024年第17卷第3期

页      面：1-10页

核心收录：

学科分类：070207[理学-光学] 07[理学] 08[工学] 080203[工学-机械设计及理论] 0802[工学-机械工程] 0803[工学-光学工程] 0702[理学-物理学] 

基　　金：supported by the National Natural Science Foundation of China(62375144 and 61875092) Tianjin Foundation of Natural Science(21JCYBJC00260) Beijing-Tianjin-Hebei Basic Research Cooperation Special Program(19JCZDJC65300) 

主　　题：Optical coherence tomography saturation artifacts deep learning image inpainting. 

摘      要：Limited by the dynamic range of the detector,saturation artifacts usually occur in optical coherence tomography(OCT)imaging for high scattering *** available methods are difficult to remove saturation artifacts and restore texture completely in OCT *** proposed a deep learning-based inpainting method of saturation artifacts in this *** generation mechanism of saturation artifacts was analyzed,and experimental and simulated datasets were built based on the *** super-resolution generative adversarial networks were trained by the clear–saturated phantom image *** perfect reconstructed results of experimental zebrafish and thyroid OCT images proved its feasibility,strong generalization,and robustness.