CogAdapt: Adapting Clinical ECG Foundation Models for Wearable Cognitive Load Assessment

Mousavi, Amir; Nourbakhsh, Erfan; Sirjani, Mohammad Sadegh; Xie, Mimi; Slavin, Rocky; Neely, Leslie; Davis, John; Quarles, John

IEEE EMBS BHI 2026

CogAdapt: Adapting Clinical ECG Foundation Models for Wearable Cognitive Load Assessment

Amir Mousavi¹, Erfan Nourbakhsh¹, Mohammad Sadegh Sirjani¹, Mimi Xie¹, Rocky Slavin¹, Leslie Neely¹, John Davis¹, John Quarles¹

¹University of Texas at San Antonio (UTSA)

IEEE-EMBS International Conference on Biomedical and Health Informatics (BHI 2026)

arXiv Code (Soon) BibTeX

Introduction

The Sensor Gap

Clinical ECG foundation models expect 12-lead hospital recordings. Wearable cognitive load datasets provide noisy 3-lead signals and a different task.

Clinical 12-lead ECG-FM versus wearable 3-lead cognitive load classifier, with adapting models question in between.

Figure 1. The challenge: pretrained foundation models expect 12-lead clinical ECG, but wearable cognitive load datasets provide only 3-lead recordings.

Paper

Abstract

Assessing cognitive load continuously and at low latency would help adaptive human-computer interaction, but it remains hard because labeled data are scarce and models generalize poorly across subjects. Recent ECG foundation models are pretrained on millions of clinical diagnostic ECG recordings, yet they do not apply directly to wearable devices when the sensor configuration and the task both differ. We present CogAdapt, a framework that adapts a clinical ECG foundation model to wearable cognitive load assessment. LeadBridge is a learnable adapter that maps 3-lead wearable signals to a 12-lead-compatible representation. ProFine is a progressive fine-tuning strategy that unfreezes encoder layers in stages while limiting representational drift. On CLARE and CL-Drive under leave-one-subject-out cross-validation, CogAdapt reaches macro-F1 of 0.626 and 0.768, improving over from-scratch baselines by 11.2 and 16.1 percentage points.

Method

CogAdapt Pipeline

Wearable preprocessing, LeadBridge 3→12 mapping, pretrained ECG-FM, and a binary load head.

Figure 2. The CogAdapt pipeline: LeadBridge (3→12 leads), the pretrained ECG-FM encoder, and ProFine fine-tuning for cognitive-load classification.

Adaptation

ProFine Progressive Fine-Tuning

Three scenarios control how much of ECG-FM is updated: frozen, top layers, or full encoder with bucketed learning rates.

ProFine scenarios A, B, and C showing frozen versus trainable LeadBridge, ECG-FM, and classification head.

Figure 3. Progressive fine-tuning scenarios. Scenario A freezes the encoder. Scenario B unfreezes top layers. Scenario C unfreezes all layers.

Results

Main Results (Table I)

Performance on CLARE and CL-Drive under K-fold and LOSO. Cells are mean ± std over folds. Best per column in bold.

Method	CLARE						CL-Drive
	K-Fold			LOSO			K-Fold			LOSO
	Acc ± std	F1 ± std	AUC	Acc ± std	F1 ± std	AUC	Acc ± std	F1 ± std	AUC	Acc ± std	F1 ± std	AUC
ECG-LightCNN	.774±.007	.726±.009	.832	.735±.136	.514±.044	.539	.814±.017	.808±.018	.887	.716±.114	.607±.093	.707
Transformer	.706±.012	.643±.011	.718	.640±.116	.496±.044	.506	.688±.014	.680±.013	.731	.617±.102	.541±.086	.604
HRV-RF	.756±.011	.688±.016	.795	.668±.167	.444±.065	.517	.768±.013	.758±.014	.835	.594±.137	.533±.123	.672
CogAdapt-A	.742±.012	.717±.010	.843	.681±.122	.527±.052	.598	.823±.018	.821±.018	.906	.665±.166	.578±.149	.705
CogAdapt-B	.762±.012	.736±.012	.857	.691±.127	.557±.085	.706	.833±.019	.831±.019	.919	.770±.073	.704±.099	.811
CogAdapt-C	.813±.033	.785±.034	.898	.736±.104	.626±.119	.799	.862±.025	.860±.025	.945	.831±.091	.768±.117	.889

Table I. Performance on CLARE and CL-Drive under K-fold and LOSO. Cells are mean ± std over folds (K-fold 10; LOSO 20 on CLARE, 21 on CL-Drive). AUC is AUROC. Best per column in bold.

Ablation

LeadBridge Ablation (Table II)

Frozen encoder (Scenario A). Only the 3→12 mapping and head are trained. LeadBridge wins on macro-F1 for both datasets.

3→12 Mapping	CLARE		CL-Drive
3→12 Mapping	F1	AUC	F1	AUC
Zero-padding	.380	.639	.519	.764
Random adapter	.409	.548	.516	.704
Dower transform	.407	.628	.508	.744
LeadBridge	.527	.598	.578	.705

Table II. LeadBridge ablation under the frozen ECG-FM encoder. Best F1 cells in bold.

Pretraining

PTB-XL Reconstruction (Table III)

Held-out PTB-XL reconstruction for precordial leads V2–V6. LeadBridge leads on RMSE for V2–V5 and on correlation for most leads.

Lead	RMSE ↓ (µV)			Correlation ↑
Lead	Dower	Lin. Reg.	LeadBridge	Dower	Lin. Reg.	LeadBridge
V2	205.69	195.08	175.61	0.546	0.546	0.669
V3	290.39	194.52	189.77	0.569	0.570	0.613
V4	270.63	164.72	164.02	0.662	0.698	0.718
V5	215.06	143.45	140.76	0.679	0.694	0.717
V6	161.40	129.66	135.22	0.544	0.549	0.547

Table III. Held-out PTB-XL reconstruction for V2–V6. RMSE in µV.

Summary

Takeaways

Clinical ECG pretraining can transfer to wearable cognitive load when the sensor gap is closed with a learned lead adapter.
LeadBridge beats zero-padding, random adapters, and fixed Dower transforms on LOSO macro-F1 under a frozen encoder.
Progressive unfreezing (ProFine) improves monotonically from A→B→C; full adaptation yields the strongest subject-independent results.
Cross-dataset transfer still drops, so environment and label physiology remain open challenges.

BibTeX

@misc{mousavi2026cogadapt,
  title={CogAdapt: Adapting Clinical ECG Foundation Models for Wearable Cognitive Load Assessment},
  author={Amir Mousavi and Erfan Nourbakhsh and Mohammad Sadegh Sirjani and Mimi Xie and Rocky Slavin and Leslie Neely and John Davis and John Quarles},
  year={2026},
  eprint={2605.22774},
  archivePrefix={arXiv},
  primaryClass={cs.LG},
  url={https://arxiv.org/abs/2605.22774},
  note={Accepted at IEEE EMBS BHI 2026},
}