Fig. 1: Overview of the generative deep learning approach: a pipeline for generating synthetically enhanced SDOCT enhanced data.
From: Optical coherence tomography choroidal enhancement using generative deep learning
a All the OCT images underwent a paired pre-processing step before being deployed in the DL model to match volume size and field of view between the two devices, align the retina, and register the image pairs. b During the training step, paired SDOCT and SSOCT data were used and the deep-learning model learned the deep anatomical features in the image from SSOCT scans. c During the testing step, SDOCT images were inputted into the trained deep learning model and synthetically enhanced SDOCT data was generated. SDOCT Spectral-Domain Optical Coherence Tomography, SSOCT Swept-Source Optical Coherence Tomography.
