Auto- ROP

Automatic detection of ROP Deep learning.

Retinopathy of prematurity (ROP) is an eye disease that can occur in premature (early) babies weighing less than 3 pounds at birth.
ROP occurs when abnormal blood vessels develop in the retina (the light-sensitive layer of tissue in the back of your eye). Retinopathy of prematurity (ROP) is the leading cause of childhood blindness worldwide. Some babies with ROP have mild cases that improve without treatment. Some babies, however, require treatment to protect their vision and prevent blindness.
ROP is divided into five stages. These stages are used by doctors to determine the severity of ROP. The stages range from mild (stage 1) to severe (stage 5). (severe)
The outcome of treatment is largely dependent on early detection and intervention, including surgery.
Automated image analysis and deep learning (DL) systems for ROP have the potential to improve ROP care by (1) increasing the efficiency, accuracy, and objectivity of diagnosis and (2) facilitating quantitative disease monitoring and risk prediction.8 AI, machine learning, and DL tools have the potential to improve ROP diagnosis and management.