Monocular Retinal Depth Estimation and Joint Optic Disc and Cup Segmentation using Adversarial Networks
This work addresses glaucoma assessment by improving depth estimation and segmentation from monocular images, which is incremental as it builds on adversarial networks for a specific medical imaging task.
The paper tackled the problem of estimating depth from a single retinal image and using it for joint optic disc and cup segmentation, achieving a high average correlation coefficient of 0.92 in depth prediction, outperforming state-of-the-art methods.
One of the important parameters for the assessment of glaucoma is optic nerve head (ONH) evaluation, which usually involves depth estimation and subsequent optic disc and cup boundary extraction. Depth is usually obtained explicitly from imaging modalities like optical coherence tomography (OCT) and is very challenging to estimate depth from a single RGB image. To this end, we propose a novel method using adversarial network to predict depth map from a single image. The proposed depth estimation technique is trained and evaluated using individual retinal images from INSPIRE-stereo dataset. We obtain a very high average correlation coefficient of 0.92 upon five fold cross validation outperforming the state of the art. We then use the depth estimation process as a proxy task for joint optic disc and cup segmentation.