DT4ECG: A Dual-Task Learning Framework for ECG-Based Human Identity Recognition and Human Activity Detection

2025-02-16

AI Summary

This work addresses the dual challenges of wearable health monitoring and biometric authentication by proposing the first end-to-end multitask framework for simultaneous ECG-based individual identification and human activity recognition. Methodologically, we introduce a Sequence-Channel Attention (SCA) mechanism to jointly model temporal dynamics and inter-channel dependencies, and employ GradNorm to adaptively balance task-specific gradients, thereby mitigating optimization conflicts between identity classification and activity recognition. The architecture adopts a 1D residual CNN backbone integrated with SCA and GradNorm-weighted multitask loss. Evaluated on standard ECG benchmarks, our model achieves 99.12% accuracy for identity verification and 90.11% for activity classification—substantially outperforming both single-task baselines and existing joint-learning approaches. The key contributions are the novel SCA mechanism and the first empirical validation of GradNorm-based multitask optimization for ECG-driven biometric and activity analysis.

Technology Category

• Planning, Routing, and Scheduling: Activity and Plan Recognition
• Machine Learning: Multimodal Learning
• Computer Vision: Video Understanding & Activity Analysis

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Security and Privacy: Large-scale security measurements
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems

Abstract

This article introduces DT4ECG, an innovative dual-task learning framework for Electrocardiogram (ECG)-based human identity recognition and activity detection. The framework employs a robust one-dimensional convolutional neural network (1D-CNN) backbone integrated with residual blocks to extract discriminative ECG features. To enhance feature representation, we propose a novel Sequence Channel Attention (SCA) mechanism, which combines channel-wise and sequential context attention to prioritize informative features across both temporal and channel dimensions. Furthermore, to address gradient imbalance in multi-task learning, we integrate GradNorm, a technique that dynamically adjusts loss weights based on gradient magnitudes, ensuring balanced training across tasks. Experimental results demonstrate the superior performance of our model, achieving accuracy rates of 99.12% in ID classification and 90.11% in activity classification. These findings underscore the potential of the DT4ECG framework in enhancing security and user experience across various applications such as fitness monitoring and personalized healthcare, thereby presenting a transformative approach to integrating ECG-based biometrics in everyday technologies.

Problem

• ECG-based human identity recognition
• ECG-based activity detection
• dual-task learning framework

Innovation

• Dual-task learning framework
• Sequence Channel Attention mechanism
• GradNorm for balanced training