Change of the thickness and volume of the choroid, which can be observed and quantified from optical coherencetomography (OCT) images, is a feature of many retinal diseases, such as aged-related macular degeneration and myopicmaculopathy. In this paper, we make purposeful improvements on the U-net for segmenting the choroid of either normalor pathological myopia retina, obtaining the Bruch’s membrane (BM) and the choroidal-scleral interface (CSI). There aretwo main improvements to the U-net framework: (1) Adding a refinement residual block (RRB) to the back of eachencoder. This strengthens the recognition ability of each stage; (2) The channel attention block (CAB) is integrated withthe U-net. This enables high-level semantic information to guide the underlying details and handle the intra-classinconsistency problem. We validated our improved network on a dataset which consists of 952 OCT Bscans obtainedfrom 95 eyes from both normal subjects and patients suffering from pathological myopia. Comparing with manualsegmentation, the mean choroid thickness difference is 8μm, and the mean Dice similarity coefficient is 85.0%.
展开▼
