Saved in:
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2605.16975 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909049921994752 |
|---|---|
| author | Tang, Wei Han, Jinpei Cui, Kangning Carletti, Mattia Gustafsson, Fredrik K. Gowda, Shreyank N Palo, Patitapaban Thakur, Anshul Clifton, Lei Morel, Jean-michel Chan, Raymond H. Clifton, David A. Gu, Xiao |
| author_facet | Tang, Wei Han, Jinpei Cui, Kangning Carletti, Mattia Gustafsson, Fredrik K. Gowda, Shreyank N Palo, Patitapaban Thakur, Anshul Clifton, Lei Morel, Jean-michel Chan, Raymond H. Clifton, David A. Gu, Xiao |
| contents | Electrocardiogram (ECG) foundation models pretrained on typical diagnostic 10-second ECG segments, have demonstrated strong transferability across a range of clinical applications. However, many real-world applications produce recordings that are typically longer, and are varied in duration during inference time. These 10-second models have no built-in way to combine information across time. Extending them to longer horizons introduces two challenges: structural incompatibilities arising from input-length disparities, and semantic challenges that limit meaningful temporal aggregation. We propose a parameter-efficient framework that extends pretrained ECG foundation models to longer and variable-length ECGs without retraining the backbone. Guided by a frozen pretrained 10-second model, we introduce a lightweight plug-in module that extends the model in two complementary ways: (i) structurally compatible long-sequence processing and (ii) semantically informed temporal modeling. Experiments on multiple long-horizon ECG tasks, datasets, and foundation model backbones demonstrate that our method enables robust long-horizon extension from pretrained snapshot models, consistently outperforming sliding-window and pooling-based baselines with strong parameter efficiency. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_16975 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Extending Pretrained 10-Second ECG Foundation Models to Longer Horizons Tang, Wei Han, Jinpei Cui, Kangning Carletti, Mattia Gustafsson, Fredrik K. Gowda, Shreyank N Palo, Patitapaban Thakur, Anshul Clifton, Lei Morel, Jean-michel Chan, Raymond H. Clifton, David A. Gu, Xiao Machine Learning Artificial Intelligence Electrocardiogram (ECG) foundation models pretrained on typical diagnostic 10-second ECG segments, have demonstrated strong transferability across a range of clinical applications. However, many real-world applications produce recordings that are typically longer, and are varied in duration during inference time. These 10-second models have no built-in way to combine information across time. Extending them to longer horizons introduces two challenges: structural incompatibilities arising from input-length disparities, and semantic challenges that limit meaningful temporal aggregation. We propose a parameter-efficient framework that extends pretrained ECG foundation models to longer and variable-length ECGs without retraining the backbone. Guided by a frozen pretrained 10-second model, we introduce a lightweight plug-in module that extends the model in two complementary ways: (i) structurally compatible long-sequence processing and (ii) semantically informed temporal modeling. Experiments on multiple long-horizon ECG tasks, datasets, and foundation model backbones demonstrate that our method enables robust long-horizon extension from pretrained snapshot models, consistently outperforming sliding-window and pooling-based baselines with strong parameter efficiency. |
| title | Extending Pretrained 10-Second ECG Foundation Models to Longer Horizons |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2605.16975 |