Promising Artificial Intelligence–Machine Learning–Deep Learning Algorithms in Ophthalmology

Abstract

The lifestyle of modern society has changed significantly with the emergence of artificial intelligence (AI), machine learning (ML) and deep learning (DL) technologies recently. There are many innovations concerning novel automated technologies including unmanned plane, autonomous vehicles, face and image recognition, automatic speech recognition, natural language processing, bioinformatics, military settings, drug discovery and toxicology. AI is a multi-dimensional technology with various components such as advanced algorithms, ML and DL. AI, ML, and DL are expected to provide recent automated devices to ophthalmologists for early diagnosing and treating ocular disorders in the near future. In fact, AI, ML, and DL are being used in ophthalmic setting to validate the diagnosis of diseases, read images, corneal topographic mapping, and IOL calculations. Diabetic retinopathy (DR), age-related macular degeneration (AMD), and glaucoma are the three most common causes of irreversible blindness on a global scale. Ophthalmic imaging provides a way to diagnose and objectively detect the progression of a number of pathologies including DR, AMD, glaucoma and other ophthalmic disorders. There are two methods of imaging generally used as diagnostic methods in ophthalmic practice, including fundus digital photography and optical coherence tomography (OCT). Particularly, OCT has become the most widely used practical imaging modality in ophthalmology settings across developed world. Changes in population demographics and life style, extension of average lifespan, and the changing pattern of chronic diseases, such as obesity, diabetes mellitus (DM), DR, AMD, and glaucoma create a rising demand for such images. Furthermore, the limitation of availability of retina specialists and trained human graders is a major problem in many countries. Consequently, given the current population growth trends, it is inevitable that analysing of such images is time consuming, costly, and prone to human error. Therefore, the detection and treatment of DR, AMD, glaucoma and other ophthalmic disorders will be inevitable through unmanned automated applications system in the near future. Namely, these challenges may be only resolved through unmanned automated applications of novel analysis methods without the need of ophthalmologists, retina specialists and human graders. We therefore provide an overview of the potential effects of the current AI, ML and DL methods and their applications on the early detection and treatment of DR, AMD, glaucoma and other ophthalmic diseases.

Artificial Intelligence in Predicting Systemic Parameters and Diseases From Ophthalmic Imaging

“…Prediction of systemic disease from RFPs is a hotly studied topic, and seems like the logical next step, given robust existing algorithms for predicting ocular diseases (for instance, diabetic retinopathy, age-related macular degeneration, and glaucoma) from RFPs ( 82 ). Prediction of certain outcomes, such as age, gender, weight, and BMI, may not be particularly meaningful, given the ease of determination or measurement of these outcomes without a complex computer algorithm.…”

Epidemiology, health policy and public health implications of visual impairment and age-related eye diseases in mainland China

“…The diagnosis of glaucoma requires comprehensive consideration of multiple indicators, including the visual field, the angle width of the anterior chamber and changes in the optic disc and retina by experienced ophthalmologists ( 137 ). However, the extremely limited glaucoma specialists, general ophthalmologists, ophthalmic graders, eye clinics or hospitals pose a significant problem in China; this may lead to a high rate of misdiagnosis and delayed treatment ( 138 ). Therefore, promoting public screening and health education regarding glaucoma will be effective for disease control in China.…”

Harnessing artificial intelligence for ophthalmic disease diagnosis: a comparative study of CNNs and Swin transformer models

“…With the rapid advancement of Artificial Intelligence (AI) technology, an increasing number of research studies are directed towards the utilization of machine learning and deep learning methods to assist doctors in diagnosing ophthalmic diseases [4]. This paper's primary objective is to investigate the application of AI in classifying ophthalmic diseases and assess its accuracy and advantages.…”