Artificial intelligence for diabetic retinopathy screening: a review

Abstract

Diabetes is a global eye health issue. Given the rising in diabetes prevalence and ageing population, this poses significant challenge to perform diabetic retinopathy (DR) screening for these patients. Artificial intelligence (AI) using machine learning and deep learning have been adopted by various groups to develop automated DR detection algorithms. This article aims to describe the state-of-art AI DR screening technologies that have been described in the literature, some of which are already commercially available. All these technologies were designed using different training datasets and technical methodologies. Although many groups have published robust diagnostic performance of the AI algorithms for DR screening, future research is required to address several challenges, for examples medicolegal implications, ethics, and clinical deployment model in order to expedite the translation of these novel technologies into the healthcare setting.

Evidence-based Canadian guidelines for tele-retina screening for diabetic retinopathy: recommendations from the Canadian Retina Research Network (CR2N) Tele-Retina Steering Committee

“…56 Furthermore, automated screening software is not yet measured against central reading gold standards, or within groups with sufficient DR prevalence in its varying stages, and provides data on "referable DR" but not on sight-threatening or worse DR or DME in the Canadian population, and little is known as to their fit in a clinical system. 61 There is not yet existent Canadian population dataset with identified biomarkers to validate screening software tools for its population.…”

Technical and imaging factors influencing performance of deep learning systems for diabetic retinopathy

“…Recently, deep learning (DL) using convolutional neural networks (CNNs) has sparked tremendous interest in medicine 6 . In ophthalmology, many DL algorithms and systems have been reported to achieve robust performances in detecting various ocular diseases from retinal photographs [7][8][9] , especially for DR [10][11][12][13] . Despite substantial promise of DL technology, it is unclear what factors may influence the performance of a DL algorithm 14 .…”

p38 promoted retinal micro-angiogenesis through up-regulated RUNX1 expression in diabetic retinopathy

“…Diabetic retinopathy (DR) is the most common microvascular complication of diabetes and remains as a leading cause of vision loss in adults [1][2][3]. There are no obvious early symptoms in diabetic retinopathy [2]; hence, when visual problems begin, retinopathy has already advanced to almost a point of no return [4].…”

“…Diabetic retinopathy (DR) is the most common microvascular complication of diabetes and remains as a leading cause of vision loss in adults [1][2][3]. There are no obvious early symptoms in diabetic retinopathy [2]; hence, when visual problems begin, retinopathy has already advanced to almost a point of no return [4]. Laser therapy is the mainly effective therapy for preservation of sight in proliferative retinopathy [5]; however, it is not effective for reversing visual loss [5].…”