Related Experiment Video
Updated: Jun 29, 2026
Electroretinogram Recording for Infants and Children under Anesthesia to Achieve Optimal Dark Adaptation and International Standards
Published on: September 3, 2020
Retinal Image Dataset of Infants and Retinopathy of Prematurity
Juraj Timkovič1,2, Jana Nowaková3, Jan Kubíček4
1University Hospital Ostrava, Clinic of Ophthalmology, Ostrava, 708 52, Czech Republic.
Insights
This study introduces a new dataset and tools for analyzing retinopathy of prematurity (ROP), a leading cause of childhood blindness. The findings aid in developing better diagnostic models for this serious infant eye condition.
Area of Science:
- Ophthalmology
- Neonatology
- Medical Imaging
Background:
- Retinopathy of prematurity (ROP) is a significant cause of childhood blindness, affecting newborns and premature infants.
- Despite advances in neonatal care, ROP remains a critical global health concern.
- Early detection and accurate diagnosis are crucial for preventing vision loss.
Purpose of the Study:
- To present a unique, comprehensive dataset of retinal images from infants screened for ROP.
- To introduce a novel software tool (ReLeSeT) for automated retinal lesion segmentation and feature extraction.
- To provide pre-processing tools to enhance retinal features for improved ROP analysis models.
Main Methods:
- Compilation of 6,004 retinal images from 188 newborns, primarily premature infants, using three different digital retinal imaging systems.
- Development and application of the ReLeSeT software for automatic segmentation and extraction of retinal lesions.
- Implementation of pre-processing tools for feature boosting of retinal lesions and blood vessels.
Main Results:
- A unique dataset of 6,004 retinal images with associated anonymized patient data for ROP screening.
- The ReLeSeT tool successfully performs automatic retinal lesion segmentation and feature extraction.
- Published pre-processing tools facilitate the development of classification and segmentation models for ROP.
Conclusions:
- The presented dataset and tools represent a valuable resource for advancing ROP research and diagnostics.
- Automated analysis of retinal images can significantly aid in the early detection and management of ROP.
- This work supports the development of more accurate computational models for ROP screening.
Abstract:
Retinopathy of prematurity (ROP) represents a vasoproliferative disease, especially in newborns and infants, which can potentially affect and damage the vision. Despite recent advances in neonatal care and medical guidelines, ROP still remains one of the leading causes of worldwide childhood blindness. The paper presents a unique dataset of 6,004 retinal images of 188 newborns, most of whom are premature infants. The dataset is accompanied by the anonymized patients' information from the ROP screening acquired at the University Hospital Ostrava, Czech Republic. Three digital retinal imaging camera systems are used in the study: Clarity RetCam 3, Natus RetCam Envision, and Phoenix ICON. The study is enriched by the software tool ReLeSeT which is aimed at automatic retinal lesion segmentation and extraction from retinal images. Consequently, this tool enables computing geometric and intensity features of retinal lesions. Also, we publish a set of pre-processing tools for feature boosting of retinal lesions and retinal blood vessels for building classification and segmentation models in ROP analysis.
