The aim of this study is to assess the performance of two machine-learning technologies, namely, deep learning (DL) and support vector machine (SVM) algorithms, for detecting central retinal vein occlusion (CRVO) in ultrawide-field fundus images. Images from 125 CRVO patients (n=125 images) and 202 non-CRVO normal subjects (n=238 images) were included in this study. Training to construct the DL model using deep convolutional neural network algorithms was provided using ultrawide-field fundus images. The SVM uses scikit-learn library with a radial basis function kernel. The diagnostic abilities of DL and the SVM were compared by assessing their sensitivity, specificity, and area under the curve (AUC) of the receiver operating characteristic curve for CRVO. For diagnosing CRVO, the DL model had a sensitivity of 98.4% (95% confidence interval (CI), 94.3–99.8%) and a specificity of 97.9% (95% CI, 94.6–99.1%) with an AUC of 0.989 (95% CI, 0.980–0.999). In contrast, the SVM model had a sensitivity of 84.0% (95% CI, 76.3–89.3%) and a specificity of 87.5% (95% CI, 82.7–91.1%) with an AUC of 0.895 (95% CI, 0.859–0.931). Thus, the DL model outperformed the SVM model in all indices assessed (P<0.001 for all). Our data suggest that a DL model derived using ultrawide-field fundus images could distinguish between normal and CRVO images with a high level of accuracy and that automatic CRVO detection in ultrawide-field fundus ophthalmoscopy is possible. This proposed DL-based model can also be used in ultrawide-field fundus ophthalmoscopy to accurately diagnose CRVO and improve medical care in remote locations where it is difficult for patients to attend an ophthalmic medical center.

With recent development of spectral-domain optical coherence tomography (SD-OCT), the pathological changes of retina can be observed in much greater detail. SD-OCT clearly delineates three highly reflective lines in the outer retina, which are external limiting membrane (ELM), photoreceptor inner and outer segment (IS/OS) junction, and cone outer segment tips (COST) in order from inside. These lines can serve as hallmarks for the evaluation of photoreceptor condition. In retinitis pigmentosa (RP) leading to photoreceptor degeneration, the ELM, IS/OS, and COST lines are shortened with the progression of the disease. In addition, shortening of the ELM, IS/OS and COST lines is significantly associated with each other. The line length is longest in the ELM, followed by the IS/OS, and COST, suggesting that retinal layer becomes disorganized first at the COST, followed by the IS/OS and finally the ELM. This finding is consistent with the previous report that the earliest histopathological change in RP is a shortening of the photoreceptor outer segments. On the other hand, retinal layer becomes restored first at the ELM, followed by the IS/OS and finally the COST after macular hole surgery. There may be a directionality of photoreceptor impairment or restoration on optical coherence tomographic image.

Evaluating the discrimination ability of a deep convolution neural network for ultrawide-field pseudocolor imaging and ultrawide-field autofluorescence of retinitis pigmentosa. In total, the 373 ultrawide-field pseudocolor and ultrawide-field autofluorescence images (150, retinitis pigmentosa; 223, normal) obtained from the patients who visited the Department of Ophthalmology, Tsukazaki Hospital were used. Training with a convolutional neural network on these learning data objects was conducted. We examined the K-fold cross validation (K = 5). The mean area under the curve of the ultrawide-field pseudocolor group was 0.998 (95% confidence interval (CI) [0.9953–1.0]) and that of the ultrawide-field autofluorescence group was 1.0 (95% CI [0.9994–1.0]). The sensitivity and specificity of the ultrawide-field pseudocolor group were 99.3% (95% CI [96.3%–100.0%]) and 99.1% (95% CI [96.1%–99.7%]), and those of the ultrawide-field autofluorescence group were 100% (95% CI [97.6%–100%]) and 99.5% (95% CI [96.8%–99.9%]), respectively. Heatmaps were in accordance with the clinician’s observations. Using the proposed deep neural network model, retinitis pigmentosa can be distinguished from healthy eyes with high sensitivity and specificity on ultrawide-field pseudocolor and ultrawide-field autofluorescence images.

Optical coherence tomography parameters predictive of visual outcome after anti-VEGF therapy for retinal vein occlusion Akiko Fujihara-Mino, Yoshinori Mitamura, Naoki Inomoto, Hiroki Sano, Kei Akaiwa, Kentaro Semba Department of Ophthalmology, Institute of Biomedical Sciences, Tokushima University Graduate School, Tokushima, Japan Purpose: To determine the optical coherence tomography (OCT) parameters that are predictive of visual outcome after anti-VEGF therapy for a retinal vein occlusion (RVO). Methods: Fifty-seven eyes with macular edema (ME) secondary to a central or branch RVO treated with bevacizumab or ranibizumab were studied. Spectral-domain OCT and microperimetry were performed before, 1, 3, and 6 months after the treatment and at the final visit. Central retinal thickness (CRT), macular volume (MV), integrity of the external limiting membrane (ELM), ellipsoid zone (EZ), and foveal bulge (FB), and photoreceptor outer segment (PROS) length were determined. Results: The mean follow-up period was 17.8±11.5 months. In 46 of the 57 eyes, a resolution of the ME was achieved. The pretreatment CRT and MV, presence of intact ELM, EZ, and FB, and PROS length at the time of ME resolution were significantly correlated with the best-corrected visual acuity and retinal sensitivity at the final visit (P<0.050). Multiple regression analyses showed that the pretreatment MV had the highest correlation with the posttreatment best-corrected visual acuity and retinal sensitivity (P<0.050). Conclusion: The CRT, MV, ELM, EZ, FB, and PROS length are predictive factors for the visual outcome after anti-VEGF therapy for RVO. Keywords: anti-VEGF therapy, retinal sensitivity, retinal vein occlusion, spectral-domain optical coherence tomography, visual acuity