Automatic diagnosis of pediatric high myopia via Attention-based Patch Residual Shrinkage network

• Published in: Expert systems with applications 2024-12, Vol.255, p.124704, Article 124704
• Main Authors: Peng, Haoran, Li, Jianqiang, Cheng, Wenxiu, Zhao, Linna, Guan, Yu, Li, Zhaosheng, Li, Li, Xu, Xi
• Format: Article
• Language: English
• Subjects: Computer-aided diagnosis; Deep learning; Fundus images; Pediatric high myopia
• ISSN: 0957-4174
• DOI: 10.1016/j.eswa.2024.124704

Fundus images can be obtained non-invasively and be adopted to monitor the follow-up on various fundus diseases, such as high myopia. Therefore, the use of fundus images for the early screening of eye diseases has principal clinical significance. However, AI-based medical research continues to face two main challenges, lack of prior-knowledge guidance and complex fundus information. In this paper, we propose a pediatric high myopia classification model, the Attention-based Patch Residual Shrinkage network (APRSnet), which facilitates early clinical diagnosis by evaluating the importance of different image features. A collection of 2492 high-resolution fundus images of children, including 768 images of high myopia and 1724 images of non-high myopia, is fed to APRSnet as the training dataset. To better understand how different features impact the classification, APRSnet is tested on multiple feature-enhanced fusion datasets. The result shows that, by removing irrelevant information in fundus images, APRSnet achieves an accuracy of 0.959 and an F1 score of 0.946 and outperforms all classic image classification networks we compared with. We also testify that, in fundus images, gradient information benefits severity classification the most since it helps with model convergence more than luminance and texture information. •An attention-based deep network is proposed for pediatric high myopia diagnosis.•We combine the strengths of both knowledge-guided and data-driven methods.•Extensive experiments are conducted to prove the effectiveness of our method.