Purpose. Although optical coherence tomography (OCT) is essential for ophthalmologists, reading of findings requires expertise. The purpose of this study is to test deep learning with image augmentation for automated detection of chorioretinal diseases. Methods. A retina specialist diagnosed 1,200 OCT images. The diagnoses involved normal eyes (n=570) and those with wet age-related macular degeneration (AMD) (n=136), diabetic retinopathy (DR) (n=104), epiretinal membranes (ERMs) (n=90), and another 19 diseases. Among them, 1,100 images were used for deep learning training, augmented to 59,400 by horizontal flipping, rotation, and translation. The remaining 100 images were used to evaluate the trained convolutional neural network (CNN) model. Results. Automated disease detection showed that the first candidate disease corresponded to the doctor’s decision in 83 (83%) images and the second candidate disease in seven (7%) images. The precision and recall of the CNN model were 0.85 and 0.97 for normal eyes, 1.00 and 0.77 for wet AMD, 0.78 and 1.00 for DR, and 0.75 and 0.75 for ERMs, respectively. Some of rare diseases such as Vogt–Koyanagi–Harada disease were correctly detected by image augmentation in the CNN training. Conclusion. Automated detection of macular diseases from OCT images might be feasible using the CNN model. Image augmentation might be effective to compensate for a small image number for training.

The technologies of ocular imaging modalities such as optical coherence tomography (OCT) and OCT angiography (OCTA) have progressed remarkably. Of these in vivo imaging modalities, recently advanced OCT technology provides high-resolution images, e.g., histologic imaging, enabling anatomical analysis of each retinal layer, including the photoreceptor layers. Recently developed OCTA also visualizes the vascular networks three-dimensionally, which provides better understanding of the retinal deep capillary layer. In addition, ex vivo analysis using autologous aqueous or vitreous humor shows that inflammatory cytokine levels including vascular endothelial growth factor (VEGF) are elevated and correlated with the severity of macular edema (ME) in eyes with retinal vein occlusion (RVO). Furthermore, a combination of multiple modalities enables deeper understanding of the pathology. Regarding therapy, intravitreal injection of anti-VEGF drugs provides rapid resolution of ME and much better visual improvements than conventional treatments in eyes with RVO. Thus, the technologies of examination and treatment for managing eyes with RVO have progressed rapidly. In this paper, we review the multimodal imaging and therapeutic strategies for eyes with RVO with the hope that it provides better understanding of the pathology and leads to the development of new therapies.

Vitreoretinal diseases are refractory to both topical and systemic pharmacological approaches because of specific environment of the eye. That is, the cornea, the sclera, nasolacrimal drainage of tears, frontward stream of aqueous humor, blood-aqueous barrier, and blood-retinal barrier strictly limit penetration and diffusion of drug into the retina. However, recent advances in intraocular drug delivery systems (DDS) have enabled drug to be delivered effectively into the eye. Clinically successful or promising cases involve non-biodegradable implants and inserts, biodegradable inserts and microparticles, intravitreal or sub-Tenon's injection of triamcinolone acetonide, and a photodynamic therapy (PDT) with verteporfin, a photosensitizer. More recently, a variety of pharmacological challenges to treat exudative age-related macular degeneration and macular edema are proceeding into clinical trials, as soon as antivascular endothelial growth factor (anti-VEGF) therapies have been proved to be effective by repeated intravitreal injections. In the near future, DDS must be required not only to develop a new treatment modality but also to improve efficacy and/or reduce injection numbers of currently available drugs. Here we introduce controlled release of drug and discussion of recent patents with biodegradable or non-biodegradable implants and drug targeting by modification of systemically administered drug