arXiv 2412.097582025-05-16cs.LGeess.SP

面向生理信号多模态可穿戴传感的基础模型研究

Toward Foundation Model for Multivariate Wearable Sensing of Physiological Signals

Halıcıoğlu Data Science Institute, University of California San Diego（哈利奇奥格鲁数据科学研究所，加州大学圣地亚哥分校）；Schools of Engineering, Arizona State University（工程学院，亚利桑那州立大学）

Yunfei Luo, Yuliang Chen, Asif Salekin, Tauhidur Rahman

备注：: The code is available at: http://github.com/Mobile-Sensing-and-UbiComp-Laboratory/NormWear

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对可穿戴生理传感数据异质性强、泛化表示难的问题，提出带通道感知注意力的多模态基础模型NormWear，在11个数据集18项健康任务中多设置下均优于基线。

摘要：

时间序列基础模型通过利用信息丰富的波形表示，在各类数据的预测等任务中表现出色。然而，可穿戴传感数据因模式和频段的差异性带来了独特挑战，尤其是在医疗健康相关结果预测方面。主要障碍在于构建可高效适配异质传感配置与应用场景的泛化表示。为解决这一问题，我们提出NormWear——首个多模态通用基础模型，旨在从可穿戴传感数据中提取泛化且信息丰富的表示。具体而言，我们设计了带有共享特殊联络[CLS] token的通道感知注意力机制，用于检测传感器内和跨传感器的信号模式，帮助模型同时结合时间序列本身及输入传感器间的关系提取更有价值的信息，使其能广泛兼容各类传感器设置。NormWear在来自多个公开数据集的多种生理信号（包括PPG、ECG、EEG、GSR和IMU）上完成预训练。我们的模型在11个公开可穿戴传感数据集上展现出卓越的泛化能力，覆盖心理健康、身体状态推断、生命体征估计和疾病风险评估领域的18项应用。在零样本、部分样本和全样本设置下，它始终优于有竞争力的基线模型，表明其在真实世界健康应用中具有广泛适用性。

英文摘要：

Time-series foundation models excel at tasks like forecasting across diverse data types by leveraging informative waveform representations. Wearable sensing data, however, pose unique challenges due to their variability in patterns and frequency bands, especially for healthcare-related outcomes. The main obstacle lies in crafting generalizable representations that adapt efficiently across heterogeneous sensing configurations and applications. To address this, we propose NormWear, the first multi-modal and ubiquitous foundation model designed to extract generalized and informative representations from wearable sensing data. Specifically, we design a channel-aware attention mechanism with a shared special liaison [CLS] token to detect signal patterns in both intra-sensor and inter-sensors. This helps the model to extract more meaningful information considering both time series themselves and the relationships between input sensors. This helps the model to be widely compatible with various sensors settings. NormWear is pretrained on a diverse set of physiological signals, including PPG, ECG, EEG, GSR, and IMU, from various public datasets. Our model shows exceptional generalizability across 11 public wearable sensing datasets, spanning 18 applications in mental health, body state inference, vital sign estimation, and disease risk evaluation. It consistently outperforms competitive baselines under zero-shot, partial-shot, and full-shot settings, indicating broad applicability in real-world health applications.

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。