Abstract
Aims To investigate the applicability of deep learning image assessment software VeriSee DR to different color fundus cameras for the screening of diabetic retinopathy (DR). Methods Color fundus images of diabetes patients taken with three different nonmydriatic fundus cameras, including 477 Topcon TRC-NW400, 459 Topcon TRC-NW8 series, and 471 Kowa nonmyd 8 series that were judged as “gradable” by one ophthalmologist were enrolled for validation. VeriSee DR was then used for the diagnosis of referable DR according to the International Clinical Diabetic Retinopathy Disease Severity Scale. Gradability, sensitivity, and specificity were calculated for each camera model. Results All images (100%) from the three camera models were gradable for VeriSee DR. The sensitivity for diagnosing referable DR in the TRC-NW400, TRC-NW8, and non-myd 8 series was 89.3%, 94.6%, and 95.7%, respectively, while the specificity was 94.2%, 90.4%, and 89.3%, respectively. Neither the sensitivity nor the specificity differed significantly between these camera models and the original camera model used for VeriSee DR development (p = 0.40, p = 0.065, respectively). Conclusions VeriSee DR was applicable to a variety of color fundus cameras with 100% agreement with ophthalmologists in terms of gradability and good sensitivity and specificity for the diagnosis of referable DR.
Highlights
Diabetic retinopathy (DR) is one of the most severe sightthreatening diseases worldwide
Color fundus images of diabetes patients taken with nonmydriatic fundus cameras, including the TRC-NW series (TRC-NW400, TRC-NW8, TRCNW8F, and TRC-NW8F plus, Topcon Inc., Japan) and the nonmyd 8 series in three general practice clinics were collected for studies
Cross-camera external validation is needed if image assessment software is applied to images of different camera models
Summary
Diabetic retinopathy (DR) is one of the most severe sightthreatening diseases worldwide. The population of patients with diabetes has been increasing in recent years; DR awareness and regular evaluation for DR at recommended time points among these individuals remain suboptimal, probably due to poor compliance and limited resources in some areas [2, 3]. The development of a cost-effective screening program for DR using fundus photography is an important issue for both patients and healthcare professionals. Various computer programs have been developed for automated analysis of color fundus images with acceptable and comparable accuracy to those of human graders [4, 5]. Despite the increased efficiency for DR screening, the software used in the automated analysis largely learns explicit disease features taught by specialists, such as the shape and number of dot hemorrhages shown on the photos, to determine DR severity. The application of specified rules to machine learning may limit the detection of undefined features that exist in retinal images
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
