Vision screening using a smartphone platform

Abstract

Objective: The main aim of this study was to evaluate the performance of a platform designed for pediatricians to screen amblyopia using a smartphone. Methods: The medical records of consecutive children who received visual screening using a smartphone platform were retrospectively reviewed. The smartphone was used with a flash concentrator case and a software for capturing images of both eyes simultaneously by a photorefraction mechanism. The platform performance was compared to the comprehensive ophthalmological examination, which is considered the gold standard for detecting amblyopia. Sensitivity, specificity, positive predictive value, and negative predictive value of the software in detecting amblyopia risk factors were calculated. Results: A total of 157 children were included, with a mean age of 6.0±.5 years (range 5–7). In 94% of the cases, the software was able to analyze the images and release a result, determining whether or not the child presented with amblyopia risk factors. Compared to the ophthalmological examination, the smartphone platform sensitivity in detecting amblyopia risk factors was 84%, the specificity was 74%, the positive predictive value was 86%, and the negative predictive value was 70%. Conclusions: The sensitivity and specificity of the smartphone photoscreening platform to detect amblyopia risk factors were within the range of traditional instrument-based vision screening technology. A smartphone photorefraction platform appears to be a promising cost-effective alternative to assist pediatricians and minimize obstacles to vision screening and amblyopia detection. Future studies are needed to gather additional comparative data.

Smartphone-Based Artificial Intelligence for the Detection and Diagnosis of Pediatric Diseases: A Comprehensive Review

“…To improve efficacy, a new version with software for capturing images of both eyes simultaneously was developed. Two studies in which this updated smartphone application has been evaluated have shown that its sensitivity and specificity to detect amblyopia risk factors were within the range of traditional instrument-based vision screening technology [ 85 , 86 ]. However, the use of a smartphone took less time, referred more children, and detected a higher number of children with amblyopia than visual acuity testing [ 87 ].…”