ABSTRACT
Introduction
In recent decades, there have been significant advances in the field of Artificial intelligence (AI), retinal imaging, and therapeutics. The specialty of retina generates huge datasets, which are ideally suited to create robust AI models for early detection, diagnosis, classification, and treatment of retinal diseases.
Areas Covered
Basic aspects of AI algorithms, machine learning models, application of AI in diabetic retinopathy (DR), retinopathy of prematurity (ROP), retinal vascular occlusion (RVO) and age-related macular degeneration (AMD) have been described, highlighting findings from important studies.
Literature Review and Methodology
Comprehensive search of indexed medical literature on Medline/PubMed and Google Scholar databases. The search terms included artificial intelligence, deep learning, machine learning in DR, AMD, ROP, retinal vascular disease, and RVO. The manuscripts published in English literature in the last two decades were selected for this review.
Expert Opinion
Several AI algorithms have been developed which are accurate and efficacious in screening, detecting, diagnosing, and aiding in managing patients with various retinal diseases. Proper external validation using large datasets and establishing their accuracy is central to increasing the confidence and acceptance of these algorithms. The application of AI to screening models can be a boon in many environments, but particularly resource-depleted settings.
Article highlights
Our article highlights the advances in the field of Artificial Intelligence, retinal imaging and therapeutics.
It provides an overview of the composition of various neural networks and the basics of AI algorithms.
It focuses on the application of artificial intelligence in the field of retinal diseases and associated imaging modalities.
It provides an overview of the impact of AI in screening of retinal diseases, and their management.
The article also highlights the pros and cons of the application of AI in the retinal diseases.
In addition, the article emphasizes the importance of establishing external validation of AI -enabled algorithms and their performance in the real world environment.
Healthcare is transformed not by mere diagnosis but by the availability of timely, accessible and affordable treatment. To achieve this, plans must be afoot to simultaneously bolster implementation of AI diagnostic algorithms as well as robust treatment networks and processes. This aspect is also highlighted in the manuscript.
Declaration of interests
The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, Honoria, stock, ownership or options, expert testimony, grants or patents received or pending, or royalties.