Artificial intelligence could help doctors determine if a person is likely to develop the eye disease glaucoma, a study has suggested.
Subscribe now for unlimited access.
$0/
(min cost $0)
or signup to continue reading
Researchers said the technology could be a "helpful diagnostic aid" for clinicians to provide more effective treatment.
Glaucoma is a group of diseases that can damage the optic nerve in the back of the eye, leading to blindness.
It can affect people at any age, but is more common in those in their 70s and 80s.
A team in Korea looked at data from the Clinical Data Warehouse (CDW) of Seoul National University Hospital Patients Research Environment, examining 12,458 eyes with suspicious early signs of glaucoma.
From this group, they focused on 210 eyes that had progress to glaucoma and 105 that had not.
All cases had been monitored every six to 12 months for at least seven years.
Researchers developed a set of predictive combinations using 15 clinical features - such as age, gender, weight and blood pressure, as well as the thickness of some parts of the eye - along with red flag signs in retinal images taken during the monitoring period.
These combinations were fed into three algorithms that automatically order and categorise data.
They said all three "performed well" and were able to predict the progression to glaucoma with an accuracy rate of between 91 and 99 per cent.
The study authors said: "We believe that with additional training and testing on a larger dataset, our (deep learning) models can be made even better, and that with such models, clinicians would be better equipped to predict individual [glaucoma suspect] patients' respective disease courses.
"Prediction of disease course on an individual-patient basis would help clinicians to present tailored management options to patients with regard to issues such as follow-up duration."
The findings were published in the British Journal of Ophthalmology.
Joanna Hodgkinson, head of research and support services at the charity Glaucoma UK, said the study "represents a meaningful step forward in the ongoing fight against this disease".
"AI technology is an exciting research area with the potential to support glaucoma diagnosis and management," she said.
"With further refinement and testing, AI could become a valuable tool to assist medical professionals in early detection and personalised patient care, offering hope to those at risk of losing their sight to glaucoma."
Australian Associated Press
