Wenxiu Cheng

Wenxiu Cheng

Auto-generated profile, based on publicly available information

Wenxiu Cheng

Faculty of Information Technology, Beijing University of Technology, Beijing, China

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
5 May 2025 in IEEE Transactions on Artificial Intelligence
Verified
Journal Article
Closed access
HPHaoran Peng
Haoran Peng
JLJianqiang Li
Jianqiang Li
WCWenxiu Cheng
Wenxiu Cheng
LZLinna Zhao
Linna Zhao
YGYu Guan
Yu Guan
YHYo-Ping Huang
Yo-Ping Huang
Retinal diseases, such as glaucoma, age-related macular degeneration, and high myopia, are major contributors to global vision loss, emphasizing the need for early detection and intervention. Current deep learning approaches for diagnosing retinal diseases using fundus images primarily focus on single-disease classification due to the scarcity and expense of diverse datasets. This limitation restricts their generalization across multiple ocular diseases and impedes transfer learning to untrained disease types. In this paper, we introduce a novel Model-Agnostic Meta-Learning framework, called Mirror Meta-Learning (MML) , which incorporates an autoencoder module to supervise the backpropagation path in few-shot learning, enhancing model initialization and adaptation. MML’s effectiveness is validated using four publicly available retinal disease binary classification datasets and a proprietary high myopia dataset. In addition, MML demonstrates robustness when tested on three well-established few-shot learning datasets. Our results show the proposed model’s superiority in terms of performance and generalizability in ocular disease classification tasks.

26 January 2024
Verified
Conference Paper
Open access
QWQuanzeng Wang
Quanzeng Wang
JLJianqiang Li
Jianqiang Li
WCWenxiu Cheng
Wenxiu Cheng
LZLinna Zhao
Linna Zhao
Pollen allergy is a common seasonal disease. Current automatic pollen classification methods have high accuracy for high-quality images. However, the imaging process of pollen can produce defocusing blur, and some pollen characteristics have subtle differences, which limits the accuracy of the methods. To address this, we introduce a supervised contrastive learning method to fine-tune the pre-trained classification network and optimize its feature representation ability. We combined multilayer scan images of the same pollen into positive example pairs, and with multilayer scan images of other classes of pollen into negative example pairs. This can learn more similar features of the same class samples and differences of different class samples. We establish a pollen dictionary based on feature representation, compare new test samples with pollen grains in the dictionary, and obtain the final pollen class through weighted voting. Experimental results show that our method is effective for distinguishing easily confused pollen grains.