Diagnostic Accuracy and Reliability of Retinal Pathology Using the Forus 3nethra Fundus Camera Compared to Ultra Wide-Field Imaging

Abstract

Purpose : The Forus 3nethra non-mydriatic fundus camera was developed as an affordable, portable screening tool for eye conditions in resource-limited settings. We evaluated the accuracy and reliability of detecting retinal pathology using Forus fundus photographs compared to clinical exam and ultra wide-field (UWF) imaging.

Methods : We recruited patients referred for fundus photography from eye clinics at Illinois Eye and Ear Infirmary. Thirty-five dilated eyes of 18 patients were included. Patients underwent mydriatic UWF imaging with the Optos 200Tx (Optos, Marlborough, MA) and fundus photography with the Forus 3nethra camera (Forus Health, Bangalore, India). Four views of each eye were taken with the 3nethra camera: central, nasal, superior, inferior. Using a web-based system, three graders independently evaluated de-identified images for the presence or absence of pathology, image clarity, and specific clinical diagnoses. The sensitivity and specificity of each grader by imaging system was determined using the clinical diagnosis by ophthalmoscopy as the reference standard. Inter-grader agreement was assessed using the unweighted kappa statistic.

Results : Prevalence of any pathology based on clinical exam was 80%; 14% choroidal lesions, 9% uveitis, and 11% diabetic retinopathy. Accuracy of detecting ocular pathology was similar between the Forus and UWF images (sensitivity 71.3% vs 77.4%, 95% CI [-28.9%, 16.7%], p=0.60; specificity 42.8% vs 47.6%, 95% CI [-56.8%, 47.2%], p=0.85) among 3 graders. Clarity of the Forus images was good or acceptable in 77% compared to 34% in UWF images (p= 0.0003). Although sample sizes were too small to determine statistical significance, there was a trend towards greater sensitivity for detection of choroidal lesions for UWF compared to Forus (93% vs 33%) but similar sensitivity for uveitis and diabetic retinopathy. Graders reported greater confidence in diagnosis with Forus compared to UWF for clinical scenarios of diabetic retinopathy (58%), uveitis (67%), and choroidal lesions (60%). Inter-grader agreement was moderate among graders for both Forus (0.50) and UWF (0.40).

Conclusions : The Forus 3nethra images demonstrated similar accuracy and reliability compared to UWF in detecting ocular pathology. Further studies will determine the utility of the Forus 3nethra as a screening tool for specific eye diseases.

This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.