Abstract
Early detection and treatment of glaucoma is of interest as it is a chronic eye disease leading to an irreversible loss of vision. Existing automated systems rely largely on fundus images for assessment of glaucoma due to their fast acquisition and cost-effectiveness. Optical Coherence Tomographic (OCT) images provide vital and unambiguous information about nerve fiber loss and optic cup morphology, which are essential for disease assessment. However, the high cost of OCT is a deterrent for deployment in screening at large scale. In this article, we present a novel CAD solution wherein both OCT and fundus modality images are leveraged to learn a model that can perform a mapping of fundus to OCT feature space. We show how this model can be subsequently used to detect glaucoma given an image from only one modality (fundus). The proposed model has been validated extensively on four public andtwo private datasets. It attained an AUC/Sensitivity value of 0.9429/0.9044 on a diverse set of 568 images, which is superior to the figures obtained by a model that is trained only on fundus features. Cross-validation was also done on nearly 1,600 images drawn from a private (OD-centric) and a public (macula-centric) dataset and the proposed model was found to outperform the state-of-the-art method by 8% (public) to 18% (private). Thus, we conclude that fundus to OCT feature space mapping is an attractive option for glaucoma detection.
- [1] . 2015. Automatic feature learning for glaucoma detection based on deep learning. Proc. MICCAI, 669–677.Google Scholar
- [2] . 2016. Convolutional neural network transfer for automated glaucoma identification. Proc. SPIE, 12th International Symposium on Medical Information Processing and Analysis.Google Scholar
- [3] . 2015. Glaucoma detection based on deep convolutional neural network. In EMBC, 715–718.Google Scholar
- [4] . 2018. G-Eyenet: A convolutional autoencoding classifier framework for the detection of glaucoma from retinal fundus images. In ICIP, 2775–2779.Google Scholar
- [5] . 2018. Performance of deep learning architectures and transfer learning for detecting glaucomatous optic neuropathy in fundus photographs. Scientific Reports (2018).Google Scholar
- [6] . 2019. Retinal image synthesis and semi-supervised learning for glaucoma assessment. IEEE Transactions on Medical Imaging 38, 9 (2019), 2211–2218.
DOI: 10.1109/TMI.2019.2903434Google Scholar - [7] . 2019. Automatic glaucoma classification using color fundus images based on convolutional neural networks and transfer learning. Biomedical Optics Express 10, 2 (2019), 892–913.Google Scholar
- [8] . 2013. Superpixel classification based optic disc and optic cup segmentation for glaucoma screening. IEEE Transactions on Medical Imaging 32 (2013), 1019–1032.Google Scholar
Cross Ref
- [9] . 2015. Robust multi-scale superpixel classification for optic cup localization. Computerized Medical Imaging and Graphics 40 (2015), 182–193.Google Scholar
- [10] . 2014. Optic cup segmentation for glaucoma detection using low-rank superpixel representation. In MICCAI, Vol 8673. 788–795.Google Scholar
- [11] . 2017. Joint optic disc and cup boundary extraction from monocular fundus images. Comput Methods Programs Biomed. 147 (2017), 51–61.Google Scholar
- [12] . 2017. Optic disc and cup segmentation methods for glaucoma detection with modification of U-Net convolutional neural network. Pattern Recognition and Image Analysis 27 (2017), 618–624. Google Scholar
Digital Library
- [13] . 2018. Joint optic disc and cup segmentation based on multi-label deep network and polar transformation. IEEE Transactions on Medical Imaging 37 (2018), 1597–1605.Google Scholar
Cross Ref
- [14] . 2016. Glaucoma classification with a fusion of segmentation and image-based features. In ISBI.Google Scholar
- [15] . 2018. Disc-aware ensemble network for glaucoma screening from fundus image. IEEE Transactions on Medical Imaging 37 (2018), 2493–2501.Google Scholar
Cross Ref
- [16] . 2011. Clinical use of OCT in assessing glaucoma progression. Ophthalmic Surgery, Lasers Imaging 42 (2011).Google Scholar
- [17] . 2015. Automated early detection of glaucoma in wavelet domain using optical coherence tomography images. Biomedical and Pharmacology Journal 8 (2015).Google Scholar
- [18] . 2018. Clinically applicable deep learning for diagnosis and referral in retinal disease. Nature Medicine 24 (2018), 1342–1350.Google Scholar
Cross Ref
- [19] . 2002. Speckle reducing anisotropic diffusion. IEEE Transactions on Image Processing 11 (2002), 1260–1270. Google Scholar
Digital Library
- [20] . 2018. Construction of a retinal atlas for macular OCT volumes. In ICIAR, 650–658.Google Scholar
- [21] . 2018. Segmentation of retinal cysts from optical coherence tomography volumes via selective enhancement. IEEE Journal of Biomedical and Health Informatics (2018).Google Scholar
- [22] . 2005. Luminosity and contrast normalization in retinal images. Medical Image Analysis (2005), 179591–179598.Google Scholar
- [23] . 2010. Robust optic disk segmentation from color retinal images. In ICVGIP, 330–336. Google Scholar
Digital Library
- [24] . 2017. Dynamic routing between capsules. In NIPS, Vol 30. Google Scholar
Digital Library
- [25] . 2018. Capsule Networks against medical imaging data challenges. arXiv:1807.07559. https://arxiv.org/abs/1807.07559.Google Scholar
- [26] . 2018. Brain tumor type classification via capsule networks. arXiv:1802.10200. https://arxiv.org/abs/1802.10200.Google Scholar
- [27] . 2016. A discriminative feature learning approach for deep face recognition. In ECCV, 499–515.Google Scholar
- [28] . 2014. Generative adversarial nets. In NIPS, 2672–2680. Google Scholar
Digital Library
- [29] . 2014. Conditional generative adversarial nets. arXiv:1411.1784. https://arxiv.org/abs/1411.1784.Google Scholar
- [30] . 2014. Drishti-GS: Retinal image dataset for optic nerve head (ONH) segmentation. In ISBI, 53–56.Google Scholar
- [31] . 2011. RIM-ONE: An open retinal image database for optic nerve evaluation. In CBMS, 1–6. Google Scholar
Digital Library
- [32] . 2018. Retinal Fundus Glaucoma Challenge. Retrieved 2019 from https://refuge.grand-challenge.org/Home.Google Scholar
- [33] . 2016. sjchoi86-HRF. Retrieved March 2019 from https://github.com/yiweichen04/retina_dataset.Google Scholar
- [34] . 2005. Estimation of the Youden index and its associated cutoff point. Biometrical Journal 47 (2005), 458–472.Google Scholar
Cross Ref
- [35] . 2018. Code for Disc-aware Ensemble Network for Glaucoma Screening from Fundus Image. Retrieved 2018 from https://github.com/HzFu/DENet_GlaucomaScreen.Google Scholar
- [36] . 2017. Grad-CAM: Visual explanations from deep networks via gradient-based localization. In IEEE International Conference on Computer Vision
(ICCV’17) , 618–626.Google ScholarCross Ref
Index Terms
Glaucoma Assessment from Fundus Images with Fundus to OCT Feature Space Mapping
Recommendations
Automated Diagnosis of Glaucoma Using Digital Fundus Images
Glaucoma is a disease of the optic nerve caused by the increase in the intraocular pressure of the eye. Glaucoma mainly affects the optic disc by increasing the cup size. It can lead to the blindness if it is not detected and treated in proper time. The ...
U-Net with Hierarchical Bottleneck Attention for Landmark Detection in Fundus Images of the Degenerated Retina
Ophthalmic Medical Image AnalysisAbstractFundus photography has routinely been used to document the presence and severity of retinal degenerative diseases such as age-related macular degeneration (AMD), glaucoma, and diabetic retinopathy (DR) in clinical practice, for which the fovea and ...
An Atomatic Fundus Image Analysis System for Clinical Diagnosis of Glaucoma
CISIS '11: Proceedings of the 2011 International Conference on Complex, Intelligent, and Software Intensive SystemsGlaucoma is a serious ocular disease and leads blindness if it couldn t be detected and treated in proper way. The diagnostic criteria for glaucoma include intraocular pressure measurement, optic nerve head evaluation, retinal nerve fiber layer and ...






Comments