AI System for Longitudinal Monitoring and Quantification of Retinal Haemorrhages to Improve the Objectivity of Retinopathy of Prematurity Screening

This project develops an AI-based clinical decision-support system for the analysis of retinal haemorrhages in preterm infants screened for retinopathy of prematurity (ROP). The system automatically detects and segments haemorrhages and quantifies their number, area, size, shape and spatial distribution. Its key feature is the longitudinal comparison of repeated examinations of the same patient, giving clinicians an objective indication of whether the findings progress, remain stable or regress. Building on AI algorithms previously developed and pilot-tested within the LERCO project, the main outcome will be the first experimental version of a clinical software environment working with different types of retinal imaging data. In the long term, the project aims at more objective and standardised ROP screening and better-timed follow-up examinations.

Principal Investigator: Jan Kubíček, Ph.D.