CogAdapt: Transferring Clinical ECG Foundation Models to Wearable Cognitive Load Assessment via Lead Adaptation

2026-05-21

AI Summary

This work addresses the challenges of scarce labeled data on wearable devices, poor cross-subject generalization, and the difficulty of directly transferring clinical ECG foundation models due to mismatched lead configurations and task discrepancies. To overcome these issues, the authors propose LeadBridge, a lead adapter that maps three-lead wearable signals into anatomically consistent 12-lead representations, combined with ProFine, a progressive fine-tuning strategy that mitigates catastrophic forgetting during transfer. This approach enables, for the first time, effective adaptation of clinical ECG foundation models to wearable-based cognitive load assessment without subject-specific calibration. Evaluated on the CLARE and CL-Drive datasets, the method achieves macro-F1 scores of 0.626 and 0.768, respectively, substantially outperforming models trained from scratch.

Abstract

Real-time cognitive load assessment is essential for adaptive human-computer interaction but remains challenging due to limited labeled data and poor cross-subject generalization. Recent ECG foundation models pre-trained on millions of clinical recordings offer rich representations, but cannot be directly applied to wearable devices due to sensor configuration mismatch and task differences. In this paper, we propose CogAdapt, a framework that adapts clinical ECG foundation models to wearable cognitive load assessment. CogAdapt introduces LeadBridge, a learnable adapter that transforms 3-lead wearable signals into anatomically consistent 12-lead representations, and ProFine, a progressive fine-tuning strategy that gradually unfreezes encoder layers while preventing catastrophic forgetting. Evaluations on two public datasets (CLARE and CL-Drive) under leave-one-subject-out cross-validation show that CogAdapt substantially outperforms baselines trained from scratch, achieving macro-F1 scores of 0.626 and 0.768. These results demonstrate the promise of foundation model adaptation for subject-independent cognitive load assessment from wearable sensors.

Problem

• cognitive load assessment
• ECG foundation models
• wearable sensors
• cross-subject generalization
• sensor configuration mismatch

Innovation

• ECG foundation model
• lead adaptation
• cognitive load assessment
• wearable sensing
• progressive fine-tuning