EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

DME-RWKV: An Interpretable Multimodal Deep Learning Framework for Predicting Anti-VEGF Response in Diabetic Macular Edema

Yan Liu iD, Xieyang Xu, Jiaying Zhang, Hui Wang, Ao Shen iD, Xuefei Song iD, Xiaofang Xu, Yao Fu

Bioengineering · 24.12.2025 · 13 (1) · 12

DOI: 10.3390/bioengineering13010012 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioengineering, online erschienen am 24.12.2025. Verfasst von Yan Liu und 7 weiteren Autor:innen. Beteiligte Einrichtung: Department of Ophthalmology, Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai 200011, China. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 32 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Key Research and Development Program of China, Bethune Charitable Foundation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Diabetic macular edema (DME) is a leading cause of vision loss, and predicting patients’ response to anti-vascular endothelial growth factor (anti-VEGF) therapy remains a clinical challenge. In this study, we developed an interpretable deep learning model for treatment prediction and biomarker analysis. We retrospectively analyzed 402 eyes from 371 patients with DME. The proposed DME-Receptance Weighted Key Value (RWKV) integrates optical coherence tomography (OCT) and ultra-widefield (UWF) imaging using Causal Attention Learning (CAL), curriculum learning, and global completion (GC) loss to enhance microlesion detection and structural consistency. The model achieved a Dice coefficient of 71.91 ± 8.50% for OCT biomarker segmentation and an AUC of 84.36% for predicting anti-VEGF response, outperforming state-of-the-art methods. By mimicking clinical reasoning with multimodal integration, DME-RWKV demonstrated strong interpretability and robustness, providing a promising AI framework for precise and explainable prediction of anti-VEGF treatment outcomes in DME.

Schlagwörter

Stichwörter der Autor:innen

causal attentioncurriculum learningdiabetic macular edemaepiretinal membranemultimodal deep learningoptical coherence tomographyultra-widefield imaging

Autor:innen und Institutionen

  • Department of Ophthalmology, Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai 200011, China.
  • Shanghai Key Laboratory of Orbital Diseases and Ocular Oncology, Shanghai 200011, China.
  • College of Information Science and Engineering, Hohai University, Changzhou 213200, China.

ORCID-Profile: Yan Liu, Ao Shen, Xuefei Song

Förderung

  • National Key Research and Development Program of China
  • Bethune Charitable Foundation

Zitieren

APA

Liu, Y., Xu, X., Zhang, J., Wang, H., Shen, A., Song, X., Xu, X., & Fu, Y. (2025). DME-RWKV: An Interpretable Multimodal Deep Learning Framework for Predicting Anti-VEGF Response in Diabetic Macular Edema. Bioengineering, 13(1), 12. https://doi.org/10.3390/bioengineering13010012

Vancouver

Liu Y, Xu X, Zhang J, Wang H, Shen A, Song X, et al. DME-RWKV: An Interpretable Multimodal Deep Learning Framework for Predicting Anti-VEGF Response in Diabetic Macular Edema. Bioengineering. 2025;13(1):12. doi:10.3390/bioengineering13010012

BibTeX
@article{Liu2025,
  author = {Liu, Yan and Xu, Xieyang and Zhang, Jiaying and Wang, Hui and Shen, Ao and Song, Xuefei and Xu, Xiaofang and Fu, Yao},
  title = {DME-RWKV: An Interpretable Multimodal Deep Learning Framework for Predicting Anti-VEGF Response in Diabetic Macular Edema},
  journal = {Bioengineering},
  year = {2025},
  volume = {13},
  number = {1},
  pages = {12},
  doi = {10.3390/bioengineering13010012}
}
Kostenlos starten →