Fig. 2From: Development and quantitative assessment of deep learning-based image enhancement for optical coherence tomographyA representative OCT image scanned from a healthy eye. A is original single-frame; B is the enhanced OCT image (Avg-5) by averaging 5 frames; C is the enhanced OCT image (DL-5) generated by deep learning method with 5 frames; D is the enhanced OCT image (Avg-50) by averaging 50 frames. Regions with red rectangle are zoomed for visual examination. As shown in (D), in order to measure the CNR, we marked manually 50 pairs of regions of interest with size 4āĆā4 pixels in different retinal tissuesBack to article page