Researchers have used artificial intelligence to provide a granular view of the risk of age-related macular degeneration (AMD) progression.
Writing in JAMA Ophthalmology, scientists describe how they used deep learning to predict an individual’s risk of developing advanced AMD over a five-year period.
The algorithm achieved an accuracy of 75.7% using the Age-Related Eye Disease Study four-step AMD classification – higher than an ophthalmologist’s accuracy rate of 73.8%.
The study used 67,401 colour fundus images from 4613 patients.
The authors concluded that deep learning has the potential to assist clinicians in longitudinal care by providing individualised, detailed risk assessments for AMD patients.
A commentary on the research published in the same edition of JAMA Ophthalmology highlighted that deep learning technology will improve with access to larger patient data sets.
“This technology does not begin and end with fundus photographs, and we look forward to an era when a near totality of patient data – clinical, imaging, genomic, and otherwise – are amenable to this powerful tool,” the authors observed.
Advancements in artificial technology may reduce the costs of managing increasing volumes of medical data and improve workflow in the clinic, they added.
“It will eventually assist with individualizing management decisions for AMD and other ophthalmic diseases worldwide,” the authors concluded.