Promising AI outcomes in diagnosing retinopathy of prematurity – Australian research

Retinopathy of prematurity (ROP) is a vasoproliferative disorder affecting the retinae of premature infants with low birthweights. ROP remains one of the leading causes of preventable childhood blindness globally despite improvements in ophthalmic care.

Geographical variation in diagnosis has also been established, with diagnosis rates of treatment-requiring ROP significantly lower in Australia and New Zealand (7.7% and 7.5%, respectively) than that of international counterparts in the United Kingdom (19.2%), Canada, and the United States (13.0%).

This Australian study describes the initial development of a AI and deep learning algorithm, ROP.AI, to automatically diagnose retinopathy of prematurity (ROP) plus disease in fundal images.

ROP.AI was trained using 6,974 fundal images from Australasian image databases. Following the identification of an optimised operating point, the algorithm diagnosed plus disease with a 97.0% sensitivity and 97.8% negative predictive value.

The authors concluded that ROP.AI is a deep learning algorithm able to automatically diagnose ROP plus disease with high sensitivity and negative predictive value. In the context of increasing global disease burden, future development may improve access to ROP diagnosis and care.

Register FREE to receive the latest news, innovations and insights from Health Industry Hub; the only one-stop-hub connecting Australia’s Pharma, MedTech and Biotech industry professionals and its key stakeholders.

Differentiate your business, build thought leadership and boost network growth by reaching the Australian Pharma, MedTech and Biotech industry professionals. Health Industry Hub offers solutions to vendors and suppliers servicing the industry.