The structure of the novel deep learning based choroidal angiography (IMAGE)
Caption
The mechanism of our proposed framework involves three components: the source branch, target branch, and adaptation branch. This framework utilizes both labeled HD B-scans and unlabeled, densely collected low-quality B-scans to train a segmentation model for choroidal sublayer segmentation
Credit
Yanye Lu, Peking University
Usage Restrictions
News organizations may use or redistribute this image, with proper attribution, as part of news coverage of this paper only.
License
Licensed content