From the Journals

Retinal AI Predicts Neonatal Lung Disease

Retinal scans already performed in the NICU could flag infants at risk for serious lung disease before standard diagnosis

Share

Deep learning models utilizing retinal images collected for retinopathy of prematurity screening have shown potential in predicting bronchopulmonary dysplasia and pulmonary hypertension in premature infants. Conducted by researchers using images from a multicenter study, the analysis revealed that a multimodal model, integrating image features and demographic factors, outperformed models relying solely on either data type. This study highlights the emerging field of oculomics, where retinal imaging may provide insights into systemic diseases, raising possibilities for early identification of cardiopulmonary conditions in vulnerable infants.

Original Source(s)

Retinal AI Predicts Neonatal Lung Disease

Retinal scans already performed in the NICU could flag infants at risk for serious lung disease before standard diagnosis

  • by Conexiant News Staff

  • February 17, 2026

  • 4 min

Related Content

Wind of Change: When to Know It’s Time to Change Jobs

PentaVision is an independent company dedicated to eye care professionals. More about our magazines, newsletters, conferences, education and other services at pentavisionmedia.com.