Related Experiment Video
Updated: Oct 12, 2025
Assessing Early Stage Open-Angle Glaucoma in Patients by Isolated-Check Visual Evoked Potential
Published on: May 25, 2020
Single-Examination Risk Prediction of Severe Retinopathy of Prematurity
Aaron S Coyner1,2, Jimmy S Chen1, Praveer Singh3,4
1Departments of Ophthalmology.
Insights
An AI model combining infant demographics and retinal imaging can identify severe retinopathy of prematurity (ROP) earlier. This approach aims to reduce infant examinations without missing critical cases of treatment-requiring ROP (TR-ROP).
Area of Science:
- Ophthalmology
- Neonatology
- Medical Artificial Intelligence
Background:
- Retinopathy of prematurity (ROP) is a primary cause of childhood blindness.
- Current ROP screening involves frequent infant eye exams, with most infants not progressing to severe disease.
- Artificial intelligence (AI) shows promise in detecting severe ROP earlier than clinical diagnosis.
Purpose of the Study:
- To develop a risk prediction model integrating AI and clinical data.
- To reduce the number of ROP examinations required for infants.
- To avoid missing cases of treatment-requiring ROP (TR-ROP).
Main Methods:
- Retinal fundus images were used to derive a vascular severity score (VSS).
- ElasticNet logistic regression models were trained using birth weight, gestational age, and VSS.
- The highest-performing model was selected based on the area under the precision-recall curve.
Main Results:
- The model combining gestational age and VSS demonstrated the best performance.
- This model achieved 100% sensitivity in identifying TR-ROP on test datasets.
- The model provided moderate to high specificity, with negative predictive values of 100%.
Conclusions:
- The developed model can identify infants with TR-ROP over a month earlier than standard diagnosis.
- This AI-driven approach can decrease the frequency of ROP screenings.
- Early detection reduces the risk of late diagnosis and treatment, minimizing infant stress.
Background And Objectives:
Retinopathy of prematurity (ROP) is a leading cause of childhood blindness. Screening and treatment reduces this risk, but requires multiple examinations of infants, most of whom will not develop severe disease. Previous work has suggested that artificial intelligence may be able to detect incident severe disease (treatment-requiring retinopathy of prematurity [TR-ROP]) before clinical diagnosis. We aimed to build a risk model that combined artificial intelligence with clinical demographics to reduce the number of examinations without missing cases of TR-ROP.
Methods:
Infants undergoing routine ROP screening examinations (1579 total eyes, 190 with TR-ROP) were recruited from 8 North American study centers. A vascular severity score (VSS) was derived from retinal fundus images obtained at 32 to 33 weeks' postmenstrual age. Seven ElasticNet logistic regression models were trained on all combinations of birth weight, gestational age, and VSS. The area under the precision-recall curve was used to identify the highest-performing model.
Results:
The gestational age + VSS model had the highest performance (mean ± SD area under the precision-recall curve: 0.35 ± 0.11). On 2 different test data sets (n = 444 and n = 132), sensitivity was 100% (positive predictive value: 28.1% and 22.6%) and specificity was 48.9% and 80.8% (negative predictive value: 100.0%).
Conclusions:
Using a single examination, this model identified all infants who developed TR-ROP, on average, >1 month before diagnosis with moderate to high specificity. This approach could lead to earlier identification of incident severe ROP, reducing late diagnosis and treatment while simultaneously reducing the number of ROP examinations and unnecessary physiologic stress for low-risk infants.
