SemPSG: A Semantic Channel-Aware Foundation Model for Polysomnography Analysis

2026-09-28

AI Summary

This study addresses the transfer learning challenges arising from heterogeneous polysomnography (PSG) channel configurations and the limitation of existing models that overlook channel physiological semantics. We propose SemPSG, a foundation model that pioneers the explicit integration of physiological semantics into signal representation and aggregation processes, thereby overcoming the constraints of conventional fixed-structure indexing. By designing a semantic-conditioned temporal encoder and a multi-view image encoder, SemPSG jointly extracts time-frequency and morphological features, enabling flexible modeling across heterogeneous data. Experimental results demonstrate that SemPSG significantly outperforms existing methods in tasks such as sleep staging and disease prediction, while exhibiting superior cross-center generalization capabilities.

Abstract

Polysomnography (PSG) integrates multiple physiological signals to provide a comprehensive characterization of human sleep, yet its heterogeneous channel configurations across centers pose substantial challenges for transferable representation learning. Existing foundation models mainly focus on physiological modeling or temporal learning, while channel identity is often treated as a fixed structural index, overlooking the physiological semantics encoded by signal modality and reference configuration. To this end, we propose SemPSG, a Semantic channel-aware foundation model for heterogeneous PSG analysis. SemPSG explicitly represents the physiological semantics of channel identity and incorporates them into both signal representation learning and channel aggregation, enabling flexible modeling across diverse data configurations. Specifically, a semantic-conditioned time-series encoder captures signal-specific temporal dynamics and cross-signal interactions, while a multi-view image encoder extracts complementary time-frequency and morphological patterns from the same physiological recordings. We evaluate SemPSG on sleep and health-related tasks, including sleep staging, sleep-disorder breathing analysis, disease prediction, cognition and emotion recognition, and demographic estimation. Extensive experiments demonstrate consistent improvements over both general-purpose time series foundation models and PSG-specific foundation models, together with generalization across heterogeneous datasets across diverse channel configurations.

Problem

• Polysomnography
• Foundation Model
• Heterogeneous Channel Configuration
• Representation Learning
• Physiological Semantics

Innovation

• Semantic Channel-Aware
• Polysomnography Foundation Model
• Heterogeneous PSG Analysis
• Multi-View Encoder
• Semantic-Conditioned Time-Series Encoding