评估基础模型在精准医学生理信号上的可迁移性
Assessing Foundation Models' Transferability to Physiological Signals in Precision Medicine
AI总结：
本文提出三阶段仿真—嵌入评估—下游验证流程，系统检验基础模型对精准医学生理信号的可迁移性，并发现 Moirai 存在特征纠缠、时序失真和场景分辨不足。
摘要：
精准医学的成功需要计算模型能够有效处理并解释异质性患者群体中的多样化生理信号。尽管基础模型已在多个领域展现出显著的迁移能力，但它们在处理对精准医学至关重要的个体特异性生理信号方面的有效性仍基本未被探索。本工作提出了一个系统流程，用于快速且高效地评估基础模型在医学情境中的迁移能力。该流程采用三阶段方法。首先，它利用生理模拟软件生成多样化且具有临床相关性的场景，尤其关注数据稀缺的医学状况。这种基于模拟的方法既支持有针对性的能力评估，也支持后续的模型微调。其次，该流程将这些模拟信号输入基础模型以获得嵌入，随后使用线性方法进行评估。该评估量化模型捕捉三个关键方面的能力：生理特征独立性、时间动态保持以及医学场景区分。最后，流程通过特定的下游医学任务验证这些表征。我们在 Moirai 时间序列基础模型上对该流程进行的初步测试揭示了其在生理信号处理中的显著局限，包括特征纠缠、时间动态扭曲以及场景区分能力下降。这些发现表明，当前基础模型在临床环境部署前可能需要大幅架构修改或针对性微调。
英文摘要：
The success of precision medicine requires computational models that can effectively process and interpret diverse physiological signals across heterogeneous patient populations. While foundation models have demonstrated remarkable transfer capabilities across various domains, their effectiveness in handling individual-specific physiological signals - crucial for precision medicine - remains largely unexplored. This work introduces a systematic pipeline for rapidly and efficiently evaluating foundation models' transfer capabilities in medical contexts. Our pipeline employs a three-stage approach. First, it leverages physiological simulation software to generate diverse, clinically relevant scenarios, particularly focusing on data-scarce medical conditions. This simulation-based approach enables both targeted capability assessment and subsequent model fine-tuning. Second, the pipeline projects these simulated signals through the foundation model to obtain embeddings, which are then evaluated using linear methods. This evaluation quantifies the model's ability to capture three critical aspects: physiological feature independence, temporal dynamics preservation, and medical scenario differentiation. Finally, the pipeline validates these representations through specific downstream medical tasks. Initial testing of our pipeline on the Moirai time series foundation model revealed significant limitations in physiological signal processing, including feature entanglement, temporal dynamics distortion, and reduced scenario discrimination. These findings suggest that current foundation models may require substantial architectural modifications or targeted fine-tuning before deployment in clinical settings.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
