Perspectives on screening in retinopathy of prematurity: new algorithms and AI tools

Helen Kearns ^1,2, Sarah Hull ^1,2

⁰Department of Ophthalmology, University of Auckland, Auckland, New Zealand.

Clinical & Experimental Optometry +

|

November 26, 2025

View abstract on PubMed

Summary

Retinopathy of prematurity (ROP) screening identifies a blinding eye disorder in premature infants. New methods using weight gain and AI may improve accuracy and reduce the burden of current screening criteria.

Area Of Science

Ophthalmology
Neonatology
Medical Imaging

Background

Retinopathy of prematurity (ROP) is a major cause of preventable childhood blindness in premature infants.
Current screening relies on birth weight and gestational age, lacking specificity and potentially missing at-risk infants.
Rising survival rates of premature infants increase the clinical and economic burden of ROP screening.

Purpose Of The Study

To review alternative ROP screening algorithms based on postnatal weight gain.
To discuss the application and limitations of these alternative algorithms.
To explore the potential of artificial intelligence (AI) in enhancing ROP screening accuracy, efficiency, and equity.

Main Methods

Review of existing literature on ROP screening criteria and alternative algorithms.
Discussion of proposed algorithms utilizing postnatal weight gain.
Exploration of AI applications in ROP screening.

Main Results

Current ROP screening criteria (birth weight, gestational age) have limitations in specificity.
Alternative algorithms using postnatal weight gain show promise but require further evaluation.
AI has the potential to significantly improve ROP screening accuracy and efficiency.

Conclusions

There is a need for more specific and efficient ROP screening methods.
Postnatal weight gain algorithms and AI offer promising avenues for improving ROP detection.
Enhanced screening strategies are crucial for reducing the burden of preventable childhood blindness globally.

Keywords:

Artificial intelligence Australasia low-middle-income countries retinopathy of prematurity screening telemedicine