arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2511.03488cs.LG
收藏

NAP:基于注意力的晚期融合用于自动睡眠分期

NAP: Attention-Based Late Fusion for Automatic Sleep Staging

Faculty of Informatics, Università della Svizzera Italiana(信息学院,瑞士意大利大学);Inst. of Digital Tech. for Personalized Healthcare, SUPSI(个性化医疗数字技术研究所,SUPSI);Sleep Wake Epilepsy Center, Department of Neurology, Inselspital University of Bern(睡眠觉醒癫痫中心,神经科,伯恩大学医院);Faculty of Informatics Università della Svizzera Italiana(信息学院,瑞士意大利大学)

Alvise Dei Rossi, Julia van der Meer, Markus H. Schmidt, Claudio L. A. Bassetti, Luigi Fiorillo, Francesca Faraci

AI总结:

NAP通过三轴注意力机制聚合多个预训练单通道模型的预测,以自适应不同输入维度,在自动睡眠分期中实现最先进的零样本泛化性能。

摘要:

多导睡眠图信号具有高度异质性,在不同数据集和临床机构之间,其模态组成(如脑电图、眼电图、心电图)、通道可用性(如额叶、枕叶脑电图)以及采集协议均存在差异。现有大多数处理多导睡眠图数据的模型依赖于固定的模态或通道子集,因此未能充分利用其固有的多模态特性。为解决这一局限,我们提出了NAP(Neural Aggregator of Predictions,预测神经聚合器),这是一种基于注意力的模型,通过三轴注意力机制学习组合多个预测流,该机制能够捕捉时间、空间和预测器级别的依赖关系。NAP经过训练可适应不同的输入维度。通过聚合来自冻结的、预训练的单通道模型的输出,NAP始终优于单个预测器和简单集成方法,在多个数据集上取得了最先进的零样本泛化性能。尽管该方法是在基于多导睡眠图的自动睡眠分期背景下进行验证的,但所提出的方法可以扩展到其他多模态生理应用。

英文摘要:

Polysomnography signals are highly heterogeneous, varying in modality composition (e.g., EEG, EOG, ECG), channel availability (e.g., frontal, occipital EEG), and acquisition protocols across datasets and clinical sites. Most existing models that process polysomnography data rely on a fixed subset of modalities or channels and therefore neglect to fully exploit its inherently multimodal nature. We address this limitation by introducing NAP (Neural Aggregator of Predictions), an attention-based model which learns to combine multiple prediction streams using a tri-axial attention mechanism that captures temporal, spatial, and predictor-level dependencies. NAP is trained to adapt to different input dimensions. By aggregating outputs from frozen, pretrained single-channel models, NAP consistently outperforms individual predictors and simple ensembles, achieving state-of-the-art zero-shot generalization across multiple datasets. While demonstrated in the context of automated sleep staging from polysomnography, the proposed approach could be extended to other multimodal physiological 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。

↑