Related Experiment Video
Updated: Jan 8, 2026
Tear-Derived Exosomal miR-15a as New Diagnostic Tool for Diabetic Retinopathy
Published on: December 30, 2025
Macretina: a dataset, to support deep learning assisted retinopathy of prematurity diagnosis
Urvesh Trivedi1, Abhishek Srivastava2, Pratik Mahajan3
1Computer Science and Engineering, Indian Institute of Technology Indore, Indore, Madhya Pradesh, 453552, India. ms2304101014@iiti.ac.in.
Insights
A new dataset, Macretina, aids AI in diagnosing Retinopathy of Prematurity (ROP) in premature infants. This expert-annotated dataset supports automated screening for ROP, potentially preventing vision loss.
Area of Science:
- Ophthalmology
- Medical Imaging
- Artificial Intelligence
Background:
- Retinopathy of Prematurity (ROP) is a critical retinal disease in premature infants, necessitating early diagnosis to prevent vision loss.
- Current AI-based ROP screening systems are limited by a lack of well-annotated, diverse public datasets.
- Existing research often relies on single-center datasets, hindering the development of generalizable AI models.
Purpose of the Study:
- To introduce Macretina, a comprehensive, expert-annotated dataset for AI-driven Retinopathy of Prematurity screening.
- To facilitate the development of robust and clinically relevant AI models for automated ROP diagnosis.
- To provide a resource supporting AI tasks including classification, object detection, and segmentation for ROP features.
Main Methods:
- Curated 1432 retinal fundus images from 112 premature infants at Macretina Hospital, India.
- Utilized the 3nethra Neo wide-field retinal imaging system for image acquisition.
- Organized the dataset into three subsets (Macretina-Ridge, Macretina-OD, Macretina-BV) for specific ROP feature analysis.
Main Results:
- Evaluated dataset subsets using Deep Convolutional Neural Networks (DCNNs).
- Achieved promising results in classification, object detection, and semantic segmentation tasks.
- Demonstrated the dataset's capability to support diverse AI applications in ROP screening.
Conclusions:
- Macretina is a valuable resource for advancing AI-based ROP diagnosis.
- The dataset's diversity supports the development of clinically relevant and generalizable AI models.
- Facilitates improved screening and early detection of Retinopathy of Prematurity.
Abstract:
Retinopathy of Prematurity (ROP) is a vision-threatening retinal disease found in premature babies, where early diagnosis is very important to prevent irreversible vision loss. In recent years, several studies have been conducted on the development of reliable AI-based screening systems. However, due to the lack of well-annotated public datasets most of them have been limited to experimental research, using single-central datasets. In this study, we introduce Macretina, a comprehensive and expert-annotated dataset curated from 1432 retinal fundus images of 112 premature babies collected at Macretina Hospital, Indore, India. These images were captured using the 3nethra Neo wide-field retinal imaging system, commonly used for retinopathy of prematurity (ROP) screening. The dataset is specially designed to support AI-based automated ROP diagnosis and is organized into three subsets, each addressing a distinct pathologically relevant retinal feature for ROP screening. The three subsets are: Macretina-Ridge which supports binary classification for ridge/demarcation line detection, Macretina-OD which supports object detection for optic disc localization, and Macretina-BV which supports semantic segmentation for blood vessel analysis. We also evaluated the utility of each subset using standard Deep Convolutional Neural Networks (DCNNs), and the experiments achieved promising results across Classification, Object Detection, and Segmentation tasks. Our dataset captures a wide range of disease severity and imaging variations, making it well-suited for developing clinically relevant and generalizable AI models.
