发表机构
Faculty of Engineering, Shenzhen MSU-BIT University; International School, Beijing University of Posts and Telecommunications; School of Computer Science and Engineering, Southeast University; School of Artificial Intelligence and Computer Science, Jiangnan University（深圳北理莫斯科大学工程学院; 北京邮电大学国际学院; 东南大学计算机科学与工程学院; 江南大学人工智能与计算机学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对多模态电子病历存在模态缺失导致性能下降的问题，本文提出含四类互补提示的多模态提示学习框架，在两种缺失场景下的临床预测性能优于现有方法。
AI 中文摘要
重症监护病房（ICU）患者的准确评估对于及时临床干预和改善患者预后至关重要。多模态电子病历（EHR）包含结构化生理时间序列和纵向临床记录，可为重症监护预测提供互补信息。但在实际临床场景中，各模态可能部分观测或完全缺失，导致现有多模态模型性能大幅下降。为应对这一挑战，本文提出一种多模态提示学习框架，用于在不同模态缺失场景下开展稳健临床预测。该框架引入四类互补提示：生成式提示、缺失信号提示、缺失类型提示和时间提示。生成式提示为不可用模态构建替代潜在表示；缺失信号提示区分观测到的表示与生成的表示；缺失类型提示使模型适配不同的模态可用性配置；时间提示对经时间编码的临床序列执行特定条件的聚合。这些提示共同使模型在统一架构内捕捉感知缺失的模态内依赖关系和跨模态交互。大量实验表明，本文方法在两种缺失场景的各项评估指标上均优于现有方法； ablation（ ablation 即 ablation study，可译为 ablation 分析，即 ablation 研究）和稳健性分析进一步验证了四类提示的互补贡献，以及该框架从不完整多模态 EHR 数据中进行临床预测的有效性。
英文摘要
Accurate assessment of patients in intensive care units (ICUs) is essential for timely clinical intervention and improved patient outcomes. Multimodal electronic health records (EHRs), including structured physiological time series and longitudinal clinical notes, provide complementary information for critical care prediction. However, in real-world clinical settings, individual modalities may be partially observed or entirely unavailable, resulting in substantial performance degradation for existing multimodal models. To address this challenge, we propose a multimodal prompt-learning framework for robust clinical prediction under diverse missing-modality scenarios. The proposed framework introduces four complementary types of prompts: generative prompts, missing-signal prompts, missing-type prompts, and temporal prompts. Generative prompts construct surrogate latent representations for unavailable modalities, while missing-signal prompts distinguish observed representations from generated ones. Missing-type prompts condition the model on different modality-availability configurations, whereas temporal prompts perform condition-specific aggregation over temporally encoded clinical sequences. Together, these prompts enable the model to capture missingness-aware intramodal dependencies and cross-modal interactions within a unified architecture. Extensive experiments demonstrate that our method outperforms existing approaches across evaluation metrics on two missingness settings. Ablation and robustness analyses further verify the complementary contributions of the four prompt types and the effectiveness of the proposed framework for clinical prediction from incomplete multimodal EHR data.
Comments13 pages, 6 figures
