A new review article has been published on Optical diagnosis and photodynamic therapy We have reviewed the major publication trends in artificial intelligence (AI) research within ophthalmology over the past 20 years.
The review found that research topics have grown “exponentially” since 2015, with deep learning, optical coherence tomography (OCT) and diabetic retinopathy emerging as key themes.
The researchers highlighted that from 2010 to 2015, academic publications combining AI and ophthalmology focused on fundamental technologies such as machine learning, algorithms, OCT, and neural networks, as well as disease-specific interests such as glaucoma.
From 2016 to 2020, research focused on “more practical themes” such as segmentation and expanding the use of OCT.
From 2021 to 2024, additional important research topics were developed in the study, such as treatment monitoring.
The scientists highlighted that the thematic trends in research over two decades show that “the field’s shift from algorithm development to clinically integrated AI solutions in ophthalmology is a sign of the maturation and diversification of the research base.”
In research on AI and ophthalmology, the study authors cited a range of “prolific institutions” including the University of London (202 publications on the subject), the Medical University of Vienna, and the National University of Singapore (191 publications each).
The authors observed that despite strong growth in the research field, significant gaps remain in “real-world clinical integration, regulatory frameworks, and representation from low-resource regions.”
“This study not only maps the current intellectual frontier of AI in ophthalmology, but also identifies important avenues for future research to ensure unbiased, interpretable, and clinically translatable AI solutions in ophthalmology,” the researchers said.
