Ph.D. student, national university of singapore
Segmentation of optic nerve head tissues using deep learning to extract for glaucoma diagnosis
Glaucoma is the second largest cause of blindness in the world. Currently, there exists no permanent cure for glaucoma. We at the NUS, have developed advanced deep learning based segmentation tools to extract various tissues out of optic nerve head scans from optical coherence tomography images (OCT). With the successful extraction of these tissues, we are able to perform parametric studies on them such as studying their thickness, deformation, and biomechanical studies, etc through which are able to improve the diagnosis of glaucoma. Currently, there exists no other deep learning based tool for the studying of the optic nerve head tissues in OCT images which can be linked to improving the diagnosis of glaucoma. Besides glaucoma, the segmentation of these tissues and their subsequent parametric study can be used for diagnosis of several ocular diseases such as age related macular degeneration, peripapillary atrophy, retinopathy, etc. Our research combines the principles of image processing to enhance the quality of images, deep learning to extract tissues, biomechanical studies to extract parametric information from these tissues and clinical inference to apply these into the meaningful diagnosis. We are a team of 13 researchers aiming to improve diagnosis of glaucoma and reduce the risk of becoming blind. We have collaborators from over 4 continents working with us with over 100,000 patient records. With no cure for glaucoma, it's important for early and advanced diagnosis to make sure its progression is reduced and thus reducing the risk of becoming blind.
Abstract: Purpose: To develop a deep learning approach to digitally-stain optical coherence tomography (OCT) images of the optic nerve head (ONH). Methods: A horizontal B-scan was acquired through the center of the ONH using OCT (Spectralis) for 1 eye of each of 100 subjects (40 normal & 60 glaucoma). All images were enhanced using adaptive compensation. A custom deep learning network was then designed and trained with the compensated images to digitally stain (i.e. highlight) 6 tissue layers of the ONH. The accuracy of our algorithm was assessed (against manual segmentations) using the Dice coefficient, sensitivity, and specificity. We further studied how compensation and the number of training images affected the performance of our algorithm. Results: For images it had not yet assessed, our algorithm was able to digitally stain the retinal nerve fiber layer + prelamina, the retinal pigment epithelium, all other retinal layers, the choroid, and the peripapillary sclera and lamina cribrosa. For all tissues, the mean dice coefficient was $0.84 \pm 0.03$, the mean sensitivity $0.92 \pm 0.03$, and the mean specificity $0.99 \pm 0.00$. Our algorithm performed significantly better when compensated images were used for training. Increasing the number of images (from 10 to 40) to train our algorithm did not significantly improve performance, except for the RPE. Conclusion. Our deep learning algorithm can simultaneously stain neural and connective tissues in ONH images. Our approach offers a framework to automatically measure multiple key structural parameters of the ONH that may be critical to improve glaucoma management.
Pub.: 24 Jul '17, Pinned: 24 Aug '17
Join Sparrho today to stay on top of science
Discover, organise and share research that matters to you