Background and Objective . Glaucomatous vision loss may be preceded by an enlargement of the cup‐to‐disc ratio (CDR). We propose to develop and validate an artificial‐intelligence‐based CDR grading system that may aid in effective glaucoma‐suspect screening. Design, Setting, and Participants . 1546 disc‐centered fundus images were selected, including all 457 images from the Retinal Image Database for Optic Nerve Evaluation dataset, and images were randomly selected from the Age‐Related Eye Disease Study and Singapore Malay Eye Study to develop the system. First, a proprietary semiautomated software was used by an expert grader to quantify vertical CDR. Then, using CDR below 0.5 (nonsuspect) and CDR above 0.5 (glaucoma suspect), deep‐learning architectures were used to train and test a binary classifier system. Measurements . The binary classifier, with glaucoma suspect as positive, is measured using sensitivity, specificity, accuracy, and AUC. Results . The system achieved an accuracy of 89.67% (sensitivity, 83.33%; specificity, 93.89%; and AUC, 0.93). For external validation, the Retinal Fundus Image Database for Glaucoma Analysis dataset, which has 638 gradable quality images, was used. Here, the model achieved an accuracy of 83.54% (sensitivity, 80.11%; specificity, 84.96%; and AUC, 0.85). Conclusions . Having demonstrated an accurate and fully automated glaucoma‐suspect screening system that can be deployed on telemedicine platforms, we plan prospective trials to determine the feasibility of the system in primary‐care settings.

Diabetic Retinopathy (DR) is one of the leading causes of blindness in the United States and other high-income countries. Early detection is key to prevention, which could be achieved effectively with a fully automated screening tool performing well on clinically relevant measures in primary care settings. We have built an artificial intelligence-based tool on a cloud-based platform for large-scale screening of DR as referable or non-referable. In this paper, we aim to validate this tool built using deep learning based techniques. The cloud-based screening model was developed and tested using deep learning techniques with 88702 images from the Kaggle dataset and externally validated using 1748 high-resolution images of the retina (or fundus images) from the Messidor-2 dataset. For validation in the primary care settings, 264 images were taken prospectively from two diabetes clinics in Queens, New York. The images were uploaded to the cloud-based software for testing the automated system as compared to expert ophthalmologists’ evaluations of referable DR. Measures used were area under the curve (AUC), sensitivity, and specificity of the screening model with respect to professional graders. The screening system achieved a high sensitivity of 99.21% and a specificity of 97.59% on the Kaggle test dataset with an AUC of 0.9992. The system was also externally validated in Messidor-2, where it achieved a sensitivity of 97.63% and a specificity of 99.49% (AUC, 0.9985). On primary care data, the sensitivity was 92.3% overall (12/13 referable images are correctly identified), and overall specificity was 94.8% (233/251 non-referable images). The proposed DR screening tool achieves state-of-the-art performance among the publicly available datasets: Kaggle and Messidor-2 to the best of our knowledge. The performance on various clinically relevant measures demonstrates that the tool is suitable for screening and early diagnosis of DR in primary care settings.

Background: Age-related macular degeneration (AMD) and diabetic retinopathy (DR) are among the leading causes of blindness in the United States and other developed countries. Early detection is the key to prevention and effective treatment. We have built an artificial intelligence-based screening system which utilizes a cloud-based platform for combined large scale screening through primary care settings for early diagnosis of these diseases. Methods: iHealthScreen Inc., an independent medical software company, has developed automated AMD and DR screening systems utilizing a telemedicine platform based on deep machine learning techniques. For both diseases, we prospectively imaged both eyes of 340 unselected non-dilated subjects over 50 years of age. For DR specifically, 152 diabetic patients at New York Eye and Ear faculty retina practices, ophthalmic and primary care clinics in New York city with color fundus cameras. Following the initial review of the images, 308 images with other confounding conditions like high myopia and vascular occlusion, and poor quality were excluded, leaving 676 eligible images for AMD and DR evaluation. Three ophthalmologists evaluated each of the images, and after adjudication, the patients were determined referrable or non-referable for AMD DR. Concerning AMD, 172 were labeled referable (intermediate or late), and 504 were non-referable (no or early). Concurrently, regarding DR, 33 were referable (moderate or worse), and 643 were non-referable (none or mild). All images were uploaded to iHealthScreen’s telemedicine platform and analyzed by the automated systems for both diseases. The system performances are tested on per eye basis with sensitivity, specificity, accuracy, and kappa scores with respect to the professional graders. Results: In identifying referable DR, the system achieved a sensitivity of 97.0% and a specificity of 96.3%, and a kappa score of 0.70 on this prospective dataset. For AMD, the sensitivity was 86.6%, the specificity of 92.1%, and a kappa score of 0.76. Conclusions: The AMD and DR screening tools achieved excellent performance operating together to identify two retinal diseases prospectively in mixed datasets, demonstrating the feasibility of such tools in the early diagnosis of eye diseases. These early screening tools will help create an even more comprehensive system capable of being trained on other retinal pathologies, a goal within reach for public health deployment.

Age-related macular degeneration (AMD) is a leading cause of blindness in the developed world. In this study, we compare the performance of retinal fundus images and genetic-information-based machine learning models for the prediction of late AMD. Using data from the Age-related Eye Disease Study, we built machine learning models with various combinations of genetic, socio-demographic/clinical, and retinal image data to predict late AMD using its severity and category in a single visit, in 2, 5, and 10 years. We compared their performance in sensitivity, specificity, accuracy, and unweighted kappa. The 2-year model based on retinal image and socio-demographic (S-D) parameters achieved a sensitivity of 91.34%, specificity of 84.49% while the same for genetic and S-D-parameters-based model was 79.79% and 66.84%. For the 5-year model, the retinal image and S-D-parameters-based model also outperformed the genetic and S-D parameters-based model. The two 10-year models achieved similar sensitivities of 74.24% and 75.79%, respectively, but the retinal image and S-D-parameters-based model was otherwise superior. The retinal-image-based models were not further improved by adding genetic data. Retinal imaging and S-D data can build an excellent machine learning predictor of developing late AMD over 2–5 years; the retinal imaging model appears to be the preferred prognostic tool for efficient patient management.

Background and Purpose: Cardiovascular disease (CVD) is a leading cause of death and disability in the world. Many CVD risk prediction models have been created, but those most widely used in clinical settings have not been externally validated, a significant gap addressed herein. Methods: Using the Multi-Ethnic Study of Atherosclerosis (MESA), we have externally validated the Framingham Risk Score, ASSIGN (Assessing the cardiovascular disease risk using SIGN) risk score, Atherosclerotic Cardiovascular Disease (ASCVD) risk score, and the European SCORE model, which were selected based on popularity among clinicians and frequency of clinical use. The models were implemented in a computer program based on published algorithms, and 100 incident CVD and 100 non-incident CVD subjects from MESA were selected for testing the repeatability. The outcome in both models achieved 100% correlation. The individual model accuracy was tested by computing their sensitivity, specificity, accuracy, and C-statistics. Results: For discrimination, ASCVD showed the highest C-statistic (0.717) and SCORE the lowest (0.677). The Framingham score provided the most consistent results among the models, with a sensitivity of 69% and a specificity of 62%, but overestimated the risks as shown by the calibration results. SCORE was relatively inconsistent, with a sensitivity of 34% and specificity of 85%, for a risk threshold set at the top 20%. Conclusions: Calibration experiments showed that the transportability of the risk was generally poor in all models, with Framingham and SCORE particularly overestimating the risks for the American population There is clearly a need for better models to predict CVD risk in the multi-ethnic American population.

In this paper, we propose an automated framework to extract three parameters (the area of newly appeared, disappeared and steady drusen) for tracking the changes in the retinal fundus images over the time. Since the retinal images of different times or visits are not necessarily exactly same, instead they could have different locations, scaling, and rotated but there are common regions (i.e., pathologies). We need to find the changes in these common regions of the images. For finding the common region, we propose an image registration algorithm based on vessel geometric shape. Following the registration of two images, we detect the optic disc based on an adaptive threshold, region growing and circle findings, and drusen based on surrounding intensity ratio. Finally, we extract three parameters for tracking the changes of drusen over time. We have tested our proposed method on 22 retinal red-free fundus images, which are randomly selected from the NAT-2 dataset and manually graded the drusen by an experienced grader. The Pearson Correlation Coefficient of 0.94 is achieved for all these three parameters for tracking the changes of the drusen by our proposed method which is better than the state-of-the-art methods. We envisage that the method will help researchers in finding the correlation between the progression of drusen and many eyes and retinal diseases.