Research map / Vision-language and multimodal / Cross-modal learning
MoEA-Net: Modality-Incremental Expert Aggregation Network for Retinal Prognostic Prediction
Hua Wang, Xiaodan Zhang, Yanzhao Shi, Chengxin Zheng, Wanyu Zhang, Zhen Wang, Jianing Wang, Xiaobing Yu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add SCA expert to OCT-only baseline (ablation) | MoEA-Net baseline · AMD-OCTA SFC | F1: 0.561 -> 0.578 | Improved |
| add TDA expert to OCT-only baseline (ablation) | MoEA-Net baseline · AMD-OCTA SFC | F1: 0.561 -> 0.577 | Improved |
| add both SCA and TDA (ablation) | MoEA-Net baseline · AMD-OCTA SFC | accuracy: 0.708 -> 0.720 | Improved |
| add SCA expert to OCT-only baseline (ablation) | MoEA-Net baseline · AMD-OCTA VR | F1: 0.600 -> 0.574 | Hurt |
| add TDA expert to OCT-only baseline (ablation) | MoEA-Net baseline · AMD-OCTA VR | F1: 0.600 -> 0.558 | Hurt |
| add Fundus modality (ablation) | MoEA-Net (OCT + SCA + TDA) · AMD-OCTA SFC | accuracy: 0.720 -> 0.737 | Improved |
| add NVL modality (all three modalities) (ablation) | MoEA-Net (OCT + Fundus) · AMD-OCTA SFC | F1: 0.593 -> 0.617 | Improved |
| add NVL modality (all three modalities) (ablation) | MoEA-Net (OCT + Fundus) · AMD-OCTA VR | F1: 0.572 -> 0.631 | Improved |
| add Fundus modality (ablation) | MoEA-Net (OCT + SCA + TDA) · AMD-OCTA VR | recall: 0.621 -> 0.613 | Hurt |
| add full MoEA-Net over LLM baseline | LLaMA3-8B + BiomedCLIP · AMD-OCTA SFC | accuracy: 0.708 -> 0.752 | Improved |
| add full MoEA-Net over LLM baseline | LLaMA3-8B + BiomedCLIP · AMD-OCTA VR | accuracy: 0.536 -> 0.583 | Improved |
| use MoEA-Net instead of CD-Chat | CD-Chat · AMD-OCTA SFC | accuracy: 0.709 -> 0.752 | Improved |
Closest papers by meaning
- FOCUS: Benchmarking Retinal Model Generalization from Foundation Vision Encoders to Multimodal LLMs (NeurIPS 2026 · similarity 0.86)
- Incomplete Modality Disentangled Representation for Ophthalmic Disease Grading and Diagnosis (AAAI 2025 · similarity 0.83)
- Like an Ophthalmologist: Dynamic Selection Driven Multi-View Learning for Diabetic Retinopathy Grading (AAAI 2025 · similarity 0.83)
- Flex-MoE: Modeling Arbitrary Modality Combination via the Flexible Mixture-of-Experts (NeurIPS 2024 · similarity 0.82)
- Interpretable Vision-Language Survival Analysis with Ordinal Inductive Bias for Computational Pathology (ICLR 2025 · similarity 0.82)
- Domain Adaptive Diabetic Retinopathy Grading with Model Absence and Flowing Data (CVPR 2025 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
