Related Experiment Video
Updated: Jun 29, 2026
Using Retinal Imaging to Study Dementia
Published on: November 6, 2017
ADS-Net: attention-awareness and deep supervision based network for automatic detection of retinopathy of prematurity
Yuanyuan Peng1, Zhongyue Chen1, Weifang Zhu1
1MIPAV Lab, School of Electronics and Information Engineering, Soochow University, Suzhou, Jiangsu Province, 215006, China.
Insights
A new deep learning model, ADS-Net, effectively detects and grades retinopathy of prematurity (ROP) in premature infants. This automated system aids ophthalmologists in diagnosing this severe eye condition, improving patient outcomes.
Area of Science:
- Ophthalmology
- Medical Imaging
- Artificial Intelligence
Background:
- Retinopathy of prematurity (ROP) is a severe ocular complication in premature infants.
- Accurate ROP detection and grading are crucial for timely intervention.
- Existing methods face challenges in learning discriminative features from fundus images due to data complexity.
Purpose of the Study:
- To develop an automated system for ROP detection and grading.
- To address limitations in feature representation for medical image analysis.
- To improve the accuracy and efficiency of ROP diagnosis.
Main Methods:
- A novel Attention-Aware and Deep Supervision network (ADS-Net) was proposed.
- A Multi-Semantic Feature Aggregation (MsFA) module using self-attention was designed to handle image complexities.
- Deeply supervised loss was employed to optimize the deep model training.
Main Results:
- ADS-Net achieved a Kappa index of 0.9552 for ROP screening and 0.9037 for ROP grading on a per-image basis.
- The proposed method demonstrated superior performance compared to state-of-the-art classification networks.
- Experimental results validated the effectiveness of ADS-Net in ROP detection and grading tasks.
Conclusions:
- The ADS-Net provides a safe, objective, and cost-effective approach for ROP diagnosis.
- The integration of attention mechanisms and deep supervision enhances feature learning and model performance.
- This automated system shows significant potential to assist ophthalmologists in managing ROP.
Abstract:
Retinopathy of prematurity (ROP) is a proliferative vascular disease, which is one of the most dangerous and severe ocular complications in premature infants. Automatic ROP detection system can assist ophthalmologists in the diagnosis of ROP, which is safe, objective, and cost-effective. Unfortunately, due to the large local redundancy and the complex global dependencies in medical image processing, it is challenging to learn the discriminative representation from ROP-related fundus images. To bridge this gap, a novel attention-awareness and deep supervision based network (ADS-Net) is proposed to detect the existence of ROP (Normal or ROP) and 3-level ROP grading (Mild, Moderate, or Severe). First, to balance the problems of large local redundancy and complex global dependencies in images, we design a multi-semantic feature aggregation (MsFA) module based on self-attention mechanism to take full advantage of convolution and self-attention, generating attention-aware expressive features. Then, to solve the challenge of difficult training of deep model and further improve ROP detection performance, we propose an optimization strategy with deeply supervised loss. Finally, the proposed ADS-Net is evaluated on ROP screening and grading tasks with per-image and per-examination strategies, respectively. In terms of per-image classification pattern, the proposed ADS-Net achieves 0.9552 and 0.9037 for Kappa index in ROP screening and grading, respectively. Experimental results demonstrate that the proposed ADS-Net generally outperforms other state-of-the-art classification networks, showing the effectiveness of the proposed method.
