By 2040, about 600 million people worldwide will have diabetes.

That’s according to a recent economic analysis modeling study, published in The Lancet Digital Health, which found that the implementation of either an automated or semi-automated deep learning system for diabetic retinopathy screening could lead to significant cost savings across the world’s health systems.

For the study, researchers looked specifically at data from 39,000 patients with diabetes in a national diabetic retinopathy screening program in Singapore in 2015.

According to the study's models, diabetes patients would incur a 12-month total cost of $77 per patient when assessed by a human, while a fully automated screening process would cut this price to $66 per patient per year.  The biggest savings, however, would come from a semi-automated approach that would cut the cost to $62 per patient per year. 

Writ large, the study said, “the savings to the Singapore health system associated with switching to the semi-automated model are estimated to be $489,000, which is roughly 20% of the current annual screening cost.”

Moreover, by 2050, Singapore is projected to have 1 million people with diabetes, so according to the current figures the estimated annual savings would be $15 million.

In discussing the results, the researchers noted that the the key difference between the fully automated model and the semi-automated model was the level of follow-up care driven by each screening method's specificity.

"The fully automated model ... yields greater savings," the researchers wrote. "This is because of a higher rate of false positives, and therefore more unnecessary specialist visits, under the fully automated model. The higher costs of graders in the semi-automated model is more than offset by the lower consultation costs. However, this is ... based on the wages in Singapore, and might not apply to other settings."

For example, the report noted, a study in Scotland showed a 46% cost reduction “by replacing first level human assessment with automated assessment in a national screening program for diabetic retinopathy, (while) another study from the UK reported cost savings of 12.8% to 21%” 

The differences in cost savings between the UK and Singapore might be due to different screening program designs (three stages in the UK vs two stages in Singapore) and pricing for consultation fees and IT platforms, among other variables.