Image quality assessment of smartphone-based retinal imaging systems

Abstract

Retinal imaging is the key to detecting several vision problems before it gets worse and may lead to severe vision loss or even blindness. Therefore, ophthalmologists suggest annual eye screenings for patients with pre-existing conditions such as diabetes. During eye exams, traditional fundus cameras are used to examine the retina. However, their large size, high price, and requirement of expertise hinder their usage at every health clinic. Therefore, smartphone-based imaging systems are an emerging research area to design small and affordable biomedical imaging devices. They may enable eye screening in remote clinics even by individuals at home. Smartphone-based retinal imaging systems are portable and have more compact designs compared to fundus cameras, so their captured images are likely to be low-quality with a smaller field of view. This paper investigates the smartphone-based portable retinal imaging systems available on the market including iExaminer, Peek Retina, D-Eye, and iNview. We first generate image distortion models for each smartphone-based system to visualize their lens distortion and light reflections on captured images. Then, we compare their image quality using deep learning based neural image assessment method to determine. Based on the results, the iNview system showed better image quality with a larger field of view compared with other smartphone-based devices.