Skip to main content
. 2022 Mar 4;22(5):2016. doi: 10.3390/s22052016

Figure 1.

Figure 1

(a) Framework of the Mask R-CNN method used for retinal and choroid layer segmentation of OCT images. (b) Faster R-CNN workflow, receiving the image in the first stage to extract features at different scales, to then, generate anchors for object detection at each feature map, keeping the ones with more retinal layer information (non-background) and classifying the object inside the selected region of interest.