SleepPriorCL: Contrastive Representation Learning with Prior Knowledge-based Positive Mining and Adaptive Temperature for Sleep Staging

2021/10/15 by Hongjun Zhang, Zhang, Hongjun, Jing Wang +7
Computer Science · Engineering · Mathematics · Medicine · Neuroscience · #Artificial intelligence #Computer science #Domain (mathematical analysis) #Domain knowledge #EEG and Brain-Computer Interfaces #FOS: Computer and information sciences #FOS: Electrical engineering #False positive paradox #Ground truth #Machine Learning (cs.LG) #Machine learning #Mathematics #Natural language processing #Obstructive Sleep Apnea Research #Pattern recognition (psychology) #Representation (politics) #Signal Processing (eess.SP) #Time Series Analysis and Forecasting #cs.LG #eess.SP #electronic engineering #information engineering

paper · pdf · doi:10.48550/arxiv.2110.09966

arxiv created 2021/10/15 · openalex publication_date 2021/10/15 · arxiv updated 2021/10/20 · openalex created_date 2021/10/25 · openalex updated_date 2026/07/28

Abstract

The objective of this paper is to learn semantic representations for sleep stage classification from raw physiological time series. Although supervised methods have gained remarkable performance, they are limited in clinical situations due to the requirement of fully labeled data. Self-supervised learning (SSL) based on contrasting semantically similar (positive) and dissimilar (negative) pairs of samples have achieved promising success. However, existing SSL methods suffer the problem that many semantically similar positives are still uncovered and even treated as negatives. In this paper, we propose a novel SSL approach named SleepPriorCL to alleviate the above problem. Advances of our approach over existing SSL methods are two-fold: 1) by incorporating prior domain knowledge into the training regime of SSL, more semantically similar positives are discovered without accessing ground-truth labels; 2) via investigating the influence of the temperature in contrastive loss, an adaptive temperature mechanism for each sample according to prior domain knowledge is further proposed, leading to better performance. Extensive experiments demonstrate that our method achieves state-of-the-art performance and consistently outperforms baselines.

Citations

• Rules for Scoring Respiratory Events in Sleep: Update of the 2007 AASM Manual for the Scoring of Sleep and Associated Events
• Analysis of a sleep-dependent neuronal feedback loop: the slow-wave microcontinuity of the EEG
• PhysioBank, PhysioToolkit, and PhysioNet
• Self-supervised Visual Feature Learning with Deep Neural Networks: A Survey
• A Simple Framework for Contrastive Learning of Visual Representations
• Supervised Contrastive Learning
• Momentum Contrast for Unsupervised Visual Representation Learning
• Self-supervised Co-training for Video Representation Learning
• Uncovering the structure of clinical EEG signals with self-supervised learning

Related

• Label-Efficient Sleep Staging Using Transformers Pre-trained with Position Prediction
• Self-supervised Learning for Label-Efficient Sleep Stage Classification: A Comprehensive Evaluation
• Stanford Sleep Bench: Evaluating Polysomnography Pre-training Methods for Sleep Foundation Models
• MEASURE: Multi-scale Minimal Sufficient Representation Learning for Domain Generalization in Sleep Staging
• Self-supervised Contrastive Learning for EEG-based Sleep Staging