融合多模态数据提升癌症生存预测，但临床数据单独表现更优。
ModalSurv: Investigating opportunities and limitations of multimodal deep survival learning in prostate and bladder cancer
- 用特定投影和交叉注意力融合临床、影像、病理与基因数据。
- 前列腺癌预测C指数达0.7402（排名第一），膀胱癌为0.5740（第五名）。
- 揭示多模态模型泛化能力有限，适合研究可解释性与数据对齐问题。
精准生存预测对个性化癌症治疗至关重要。我们提出ModalSurv，一种整合临床、MRI、组织病理学和RNA测序数据的多模态深度生存学习框架，通过模态专用投影与交叉注意力融合。在CHIMERA Grand Challenge数据集上，其在前列腺癌预测中取得0.7402的C-index（排名第一），膀胱癌为0.5740（第五名）。值得注意的是，仅使用临床特征在外部测试中反而优于多模态模型，凸显了多模态对齐不足与过拟合风险。局部验证显示多模态有增益，但泛化能力有限。ModalSurv系统评估了多模态生存建模的潜力与当前局限，强调其在可扩展、泛化性癌症预后中的前景与挑战。
原文摘要 · Abstract (English)
Accurate survival prediction is essential for personalised cancer treatment. We propose ModalSurv, a multimodal deep survival framework integrating clinical, MRI, histopathology, and RNA-sequencing data via modality-specific projections and cross-attention fusion. On the CHIMERA Grand Challenge datasets, ModalSurv achieved a C-index of 0.7402 (1st) for prostate and 0.5740 (5th) for bladder cancer. Notably, clinical features alone outperformed multimodal models on external tests, highlighting challenges of limited multimodal alignment and potential overfitting. Local validation showed multimodal gains but limited generalisation. ModalSurv provides a systematic evaluation of multimodal survival modelling, underscoring both its promise and current limitations for scalable, generalisable cancer prognosis.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。