PSDNorm: Test-Time Temporal Normalization for Deep Learning in Sleep Staging

2025/03/06 by Théo Gnassounou, Antoine Collas, Gnassounou, Théo +5 · 2 citations
Computer Science · Medicine · #FOS: Computer and information sciences #Human Pose and Action Recognition #Machine Learning (cs.LG) #Obstructive Sleep Apnea Research #Speech and Audio Processing

paper · pdf · doi:10.48550/arxiv.2503.04582

openalex publication_date 2025/03/06 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Distribution shift poses a significant challenge in machine learning, particularly in biomedical applications using data collected across different subjects, institutions, and recording devices, such as sleep data. While existing normalization layers, BatchNorm, LayerNorm and InstanceNorm, help mitigate distribution shifts, when applied over the time dimension they ignore the dependencies and auto-correlation inherent to the vector coefficients they normalize. In this paper, we propose PSDNorm that leverages Monge mapping and temporal context to normalize feature maps in deep learning models for signals. Evaluations with architectures based on U-Net or transformer backbones trained on 10K subjects across 10 datasets, show that PSDNorm achieves state-of-the-art performance on unseen left-out datasets while being more robust to data scarcity.

Cited by

• Leveraging Generic Time Series Foundation Models for EEG Classification
• CauKer: Classification Time Series Foundation Models Can Be Pretrained on Synthetic Data

Related

• Convolutional Monge Mapping Normalization for learning on sleep data
• Channel-Selective Normalization for Label-Shift Robust Test-Time Adaptation
• MEASURE: Multi-scale Minimal Sufficient Representation Learning for Domain Generalization in Sleep Staging
• Discover Your Neighbors: Advanced Stable Test-Time Adaptation in Dynamic World
• U-Time: A Fully Convolutional Network for Time Series Segmentation Applied to Sleep Staging