{"ID":2877872,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.18632","arxiv_id":"2508.18632","title":"Decouple, Reorganize, and Fuse: A Multimodal Framework for Cancer Survival Prediction","abstract":"Cancer survival analysis commonly integrates information across diverse medical modalities to make survival-time predictions. Existing methods primarily focus on extracting different decoupled features of modalities and performing fusion operations such as concatenation, attention, and \\revm{Mixture-of-Experts (MoE)-based fusion. However, these methods still face two key challenges: i) Fixed fusion schemes (concatenation and attention) can lead to model over-reliance on predefined feature combinations, limiting the dynamic fusion of decoupled features; ii) in MoE-based fusion methods, each expert network handles separate decoupled features, which limits information interaction among the decoupled features. To address these challenges, we propose a novel Decoupling-Reorganization-Fusion framework (DeReF), which devises a random feature reorganization strategy between modalities decoupling and dynamic MoE fusion modules.Its advantages are: i) it increases the diversity of feature combinations and granularity, enhancing the generalization ability of the subsequent expert networks; ii) it overcomes the problem of information closure and helps expert networks better capture information among decoupled features. Additionally, we incorporate a regional cross-attention network within the modality decoupling module to improve the representation quality of decoupled features. Extensive experimental results on our in-house Liver Cancer (LC) and three widely used TCGA public datasets confirm the effectiveness of our proposed method. Codes are available at https://github.com/ZJUMAI/DeReF.","short_abstract":"Cancer survival analysis commonly integrates information across diverse medical modalities to make survival-time predictions. Existing methods primarily focus on extracting different decoupled features of modalities and performing fusion operations such as concatenation, attention, and \\revm{Mixture-of-Experts (MoE)-ba...","url_abs":"https://arxiv.org/abs/2508.18632","url_pdf":"https://arxiv.org/pdf/2508.18632v2","authors":"[\"Huayi Wang\",\"Haochao Ying\",\"Yuyang Xu\",\"Qibo Qiu\",\"Cheng Zhang\",\"Danny Z. Chen\",\"Ying Sun\",\"Jian Wu\"]","published":"2025-08-26T03:18:25Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Generative Adversarial Network\"]","has_code":false,"code_links":[{"ID":610422,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2877872,"paper_url":"https://arxiv.org/abs/2508.18632","paper_title":"Decouple, Reorganize, and Fuse: A Multimodal Framework for Cancer Survival Prediction","repo_url":"https://github.com/ZJUMAI/DeReF","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}