PhysioME: A Robust Multimodal Self-Supervised Framework for Physiological Signals with Missing Modalities

TL;DR

PhysioME tackles the challenge of missing or corrupted modalities in physiological signals by introducing a robust, single-model multimodal SSL framework. It combines a Dual-Path NeuroNet backbone with modality-specific encoders, a ViT-based multimodal encoder, and a deep restoration decoder to reconstruct missing modality tokens, guided by a joint loss that fuses intra-modality reconstruction, missing-modality reconstruction, and cross-modality contrastive objectives. The approach yields strong generalization across sleep-stage classification and hypotension prediction under various missing-modality conditions, often matching or surpassing dedicated models while maintaining stability when data are incomplete. This work demonstrates a practical, robust solution for real-world clinical settings where sensor failures and artifacts are common, with substantial implications for continuous patient monitoring and decision support.

Abstract

Missing or corrupted modalities are common in physiological signal-based medical applications owing to hardware constraints or motion artifacts. However, most existing methods assume the availability of all modalities, resulting in substantial performance degradation in the absence of any modality. To overcome this limitation, this study proposes PhysioME, a robust framework designed to ensure reliable performance under missing modality conditions. PhysioME adopts: (1) a multimodal self-supervised learning approach that combines contrastive learning with masked prediction; (2) a Dual-PathNeuroNet backbone tailored to capture the temporal dynamics of each physiological signal modality; and (3) a restoration decoder that reconstructs missing modality tokens, enabling flexible processing of incomplete inputs. The experimental results show that PhysioME achieves high consistency and generalization performance across various missing modality scenarios. These findings highlight the potential of PhysioME as a reliable tool for supporting clinical decision-making in real-world settings with imperfect data availability.

Figures (2)

• Figure 1: Overview of PhysioME architecture. (A) Training workflow of PhysioME: a multimodal SSL framework for missing modalities. (B) Inference procedure using the pretrained PhysioME.
• Figure 2: DP-NeuroNet structure.