Related Experiment Video
Updated: May 28, 2025
Author Spotlight: Enhancing PSC-to-Functional Cell Differentiation Using ML Models Based on Live-Cell Bright-Field Imaging
Published on: October 4, 2024
Classification of Stages 1,2,3 and Preplus, Plus disease of ROP using MultiCNN_LSTM classifier
Ranjana Agrawal1, Sucheta Kulkarni2, Madan Deshpande2
1Dr. Vishwanath Karad MIT World Peace University, Pune, India.
Insights
This study introduces an explainable AI system for detecting retinopathy of prematurity (ROP) in premature infants. The system accurately classifies ROP stages and identifies Plus disease, crucial for preventing infant blindness.
Area of Science:
- Ophthalmology
- Medical Imaging
- Artificial Intelligence
Background:
- Retinopathy of prematurity (ROP) is a leading cause of blindness in premature infants.
- Accurate staging and identification of Plus disease are critical for timely ROP treatment.
- Current diagnostic methods rely on manual retinal image examination, which can be subjective and time-consuming.
Purpose of the Study:
- To develop and evaluate an explainable automated system for ROP screening.
- To improve the accuracy of classifying ROP stages (1-3) and detecting Pre-plus/Plus disease.
- To leverage deep learning models for enhanced ROP diagnosis from retinal fundus images.
Main Methods:
- Utilized multiple Convolutional Neural Networks (CNNs) for feature extraction from retinal images.
- Employed Long Short-Term Memory (LSTM) networks for image classification.
- Developed and used the Cropped STAGE and HVDROPDB-PLUS datasets, including RetCam and Neo images.
Main Results:
- The proposed MultiCNN-LSTM networks demonstrated superior performance compared to individual CNNs and CNN-LSTM models.
- Achieved higher accuracy and F1 scores in classifying ROP stages and identifying Plus disease.
- The system provides explainable classifications for ROP screening.
Conclusions:
- The developed explainable AI system shows significant potential for accurate and efficient ROP screening.
- This automated approach can aid ophthalmologists in early detection and treatment of ROP.
- Further validation on larger datasets is warranted to confirm clinical utility.
Abstract:
Retinopathy of prematurity (ROP) is a retinal disorder that can cause blindness in premature infants with low birth weight. Early detection and timely treatment are crucial to prevent blindness associated with ROP. It's essential to identify the stage and presence of Plus disease accurately when examining retinal images of at-risk infants. We are developing an explainable automated ROP screening system for the HVDROPDB datasets. The fundus images were classified as without stage (Normal)/with Stage (ROP) by segmenting the ridge. Stages 1-3 were classified using machine Learning (ML) models.•This study aims to improve accuracy of Stages 1-3 classification and identify Pre-plus/ Plus disease using MultiCNN_LSTM networks. This is accomplished by using multiple CNNs (Convolutional Neural Networks) to extract features and LSTM (Long Short-Term Memory) classifier to classify images.•Cropped STAGE dataset and HVDROPDB-PLUS dataset are constructed with RetCam and Neo images.•The proposed networks outperform individual CNNs and CNN_LSTM networks in terms of accuracy and F1 score.
