A Systematic Review on Eye as a Biomarker and an Application of Quantum Neural Network

Abstract

Artificial intelligence approaches, such as deep learning models, can be utilized for image analysis tasks such as disease prediction from retinal images. In addition to eye disorders, retinal images may help identify a variety of non-eye ailments, including diabetes, hypertension, chronic kidney disease, anemia, Alzheimer's and Parkinson's disease, and other conditions associated with age and gender. This research aims to establish the eye as a biomarker to diagnose non-ocular systemic diseases by applying quantum neural networks to diagnose diabetic retinopathy using fundus retinal images. To do so, we assess past research on recognizing non-ocular disorders utilizing the power of AI algorithms and retinal images. This study presents a systematic review involving the organization, summarization, and analysis of 51 full-length articles chosen by the PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analyses) model for paper selection. Significant and intriguing discoveries indicate that retinal images are widely utilized to diagnose diseases such as diabetes and hypertension, where the thickness of the blood vessels inside the retina is dramatically altered. However, research is still ongoing to establish a baseline for some disorders. AI techniques diagnose other diseases, i.e., anemia, Alzheimer's, Parkinson's, and kidney disease. Our review shows that non-ocular diseases can be diagnosed using only retinal images. Hence, the eye is a biomarker, and, unquestionably, it is a significant discovery for medical purposes. For this, the time and cost of diagnosis could be drastically reduced, and only a retinal snapshot would be required to detect a collection of chronic diseases.