Multi-View Contrastive Learning for Robust Domain Adaptation in Medical Time Series Analysis

2025/06/27 by YongKyung Oh, Oh, YongKyung, Alex Bui +1
Computer Science · Health Professions · #Advanced Technologies in Various Fields #Artificial Intelligence (cs.AI) #Artificial Intelligence in Healthcare #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning in Healthcare

paper · pdf · doi:10.48550/arxiv.2506.22393

openalex publication_date 2025/06/27 · openalex created_date 2025/10/16 · openalex updated_date 2026/07/28

Abstract

Adapting machine learning models to medical time series across different domains remains a challenge due to complex temporal dependencies and dynamic distribution shifts. Current approaches often focus on isolated feature representations, limiting their ability to fully capture the intricate temporal dynamics necessary for robust domain adaptation. In this work, we propose a novel framework leveraging multi-view contrastive learning to integrate temporal patterns, derivative-based dynamics, and frequency-domain features. Our method employs independent encoders and a hierarchical fusion mechanism to learn feature-invariant representations that are transferable across domains while preserving temporal coherence. Extensive experiments on diverse medical datasets, including electroencephalogram (EEG), electrocardiogram (ECG), and electromyography (EMG) demonstrate that our approach significantly outperforms state-of-the-art methods in transfer learning tasks. By advancing the robustness and generalizability of machine learning models, our framework offers a practical pathway for deploying reliable AI systems in diverse healthcare settings.

Citations

• Stable Neural Stochastic Differential Equations in Analyzing Irregular Time Series Data
• Soft Contrastive Learning for Time Series
• Self-Supervised Learning for Time Series Analysis: Taxonomy, Progress, and Prospects
• Domain Adaptation for Time Series Under Feature and Label Shifts
• SimMTM: A Simple Pre-Training Framework for Masked Time-Series Modeling
• Ti-MAE: Self-Supervised Masked Time Series Autoencoders
• Contrastive Learning for Unsupervised Domain Adaptation of Time Series
• FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness
• CoST: Contrastive Learning of Disentangled Seasonal-Trend Representations for Time Series Forecasting
• Improving Multimodal Fusion with Hierarchical Mutual Information Maximization for Multimodal Sentiment Analysis
• Time-Series Representation Learning via Temporal and Contextual Contrasting
• TS2Vec: Towards Universal Representation of Time Series
• Unsupervised Representation Learning for Time Series with Temporal Neighborhood Coding
• Domain Adaptation for Time Series Forecasting via Attention Sharing
• Time Series Domain Adaptation via Sparse Associative Structure Alignment
• Exploring Contrastive Learning in Human Activity Recognition for Healthcare
• A Brief Review of Domain Adaptation
• A Transformer-based Framework for Multivariate Time Series\n Representation Learning
• Rethinking Attention with Performers
• Conv-Linformer: Boosting Linformer's Performance with Convolution in Small-Scale Settings
• CLOCS: Contrastive Learning of Cardiac Signals Across Space, Time, and Patients
• Multi-Source Deep Domain Adaptation with Weak Supervision for Time-Series Sensor Data
• A Simple Framework for Contrastive Learning of Visual Representations
• Momentum Contrast for Unsupervised Visual Representation Learning
• Momentum Contrast for Unsupervised Visual Representation Learning
• On Mutual Information Maximization for Representation Learning
• The UEA multivariate time series classification archive, 2018
• Representation Learning with Contrastive Predictive Coding
• Mixed Precision Training
• Deeper, Broader and Artier Domain Generalization
• Attention Is All You Need
• Adversarial Discriminative Domain Adaptation
• Deep CORAL: Correlation Alignment for Deep Domain Adaptation
• Learning Transferable Features with Deep Adaptation Networks
• Unsupervised Domain Adaptation by Backpropagation
• A Survey on Transfer Learning
• Analysis of a sleep-dependent neuronal feedback loop: the slow-wave microcontinuity of the EEG

Related

• Multi-Scale and Multi-Modal Contrastive Learning Network for Biomedical Time Series
• Discrepancy-Aware Contrastive Adaptation in Medical Time Series Analysis
• A Learnable Multi-views Contrastive Framework with Reconstruction Discrepancy for Medical Time-Series
• TS-P²CL: Plug-and-Play Dual Contrastive Learning for Vision-Guided Medical Time Series Classification
• CALDA: Improving Multi-Source Time Series Domain Adaptation with Contrastive Adversarial Learning