NAP: Attention-Based Late Fusion for Automatic Sleep Staging
This addresses the challenge of inconsistent modality and channel availability in sleep staging for clinical applications, though it is incremental as it builds on existing single-channel models.
The paper tackled the problem of heterogeneous polysomnography signals in sleep staging by introducing NAP, an attention-based model that aggregates predictions from single-channel models, achieving state-of-the-art zero-shot generalization across multiple datasets.
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.