# CLEF: Clinically-Guided Contrastive Learning for Electrocardiogram Foundation Models

arXiv: https://arxiv.org/abs/2512.02180 · HTML: https://mlsanity.com/paper/2512.02180/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 30 of its 80 references that are in the map.
- First posted on arXiv 1 Dec 2025.

## Builds on (most cited first, 30 of 30)

- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [Bootstrap your own latent: A new approach to self-supervised Learning](https://mlsanity.com/paper/2006.07733/) (2020)
- [Aggregated Residual Transformations for Deep Neural Networks](https://mlsanity.com/paper/1611.05431/) (2016)
- [Unsupervised Feature Learning via Non-Parametric Instance-level Discrimination](https://mlsanity.com/paper/1805.01978/) (2018)
- [Contrastive Learning with Hard Negative Samples](https://mlsanity.com/paper/2010.04592/) (2020)
- [Unified Training of Universal Time Series Forecasting Transformers](https://mlsanity.com/paper/2402.02592/) (2024)
- [MOMENT: A Family of Open Time-series Foundation Models](https://mlsanity.com/paper/2402.03885/) (2024)
- [CLOCS: Contrastive Learning of Cardiac Signals Across Space, Time, and Patients](https://mlsanity.com/paper/2005.13249/) (2020)
- [Automatic diagnosis of the 12-lead ECG using a deep neural network](https://mlsanity.com/paper/1904.01949/) (2019)
- [ECG-FM: An Open Electrocardiogram Foundation Model](https://mlsanity.com/paper/2408.05178/) (2024)
- [Guiding Masked Representation Learning to Capture Spatio-Temporal Relationship of Electrocardiogram](https://mlsanity.com/paper/2402.09450/) (2024)
- [Self-supervised representation learning from 12-lead ECG data](https://mlsanity.com/paper/2103.12676/) (2021)
- [Zero-Shot ECG Classification with Multimodal Learning and Test-time Clinical Knowledge Enhancement](https://mlsanity.com/paper/2403.06659/) (2024)
- [HOLMES: Health OnLine Model Ensemble Serving for Deep Learning Models in Intensive Care Units](https://mlsanity.com/paper/2008.04063/) (2020)
- [Lead-agnostic Self-supervised Learning for Local and Global Representations of Electrocardiogram](https://mlsanity.com/paper/2203.06889/) (2022)
- [Reading Your Heart: Learning ECG Words and Sentences via Pre-training ECG Language Model](https://mlsanity.com/paper/2502.10707/) (2025)
- [An Electrocardiogram Foundation Model Built on over 10 Million Recordings with External Evaluation across Multiple Domains](https://mlsanity.com/paper/2410.04133/) (2024)
- [LSM-2: Learning from Incomplete Wearable Sensor Data](https://mlsanity.com/paper/2506.05321/) (2025)
- [Beyond Sensor Data: Foundation Models of Behavioral Data from Wearables Improve Health Predictions](https://mlsanity.com/paper/2507.00191/) (2025)
- [Does Negative Sampling Matter? A Review with Insights into its Theory and Applications](https://mlsanity.com/paper/2402.17238/) (2024)
- [Rethinking Negative Pairs in Code Search](https://mlsanity.com/paper/2310.08069/) (2023)
- [Icentia11K: An Unsupervised Representation Learning Dataset for Arrhythmia Subtype Discovery](https://mlsanity.com/paper/1910.09570/) (2019)
- [Analysis of Augmentations for Contrastive ECG Representation Learning](https://mlsanity.com/paper/2206.07656/) (2022)
- [Metadata-guided Consistency Learning for High Content Images](https://mlsanity.com/paper/2212.11595/) (2022)
- [Geometrically Mappable Image Features](https://mlsanity.com/paper/2003.09682/) (2020)
- [MVKT-ECG: Efficient Single-lead ECG Classification on Multi-Label Arrhythmia by Multi-View Knowledge Transferring](https://mlsanity.com/paper/2301.12178/) (2023)
- [Predicting Cardiovascular Disease Risk using Photoplethysmography and Deep Learning](https://mlsanity.com/paper/2305.05648/) (2023)
- [ECG for high-throughput screening of multiple diseases: Proof-of-concept using multi-diagnosis deep learning from population-based datasets](https://mlsanity.com/paper/2210.06291/) (2022)
- [A Wearable ECG Monitor for Deep Learning Based Real-Time Cardiovascular Disease Detection](https://mlsanity.com/paper/2201.10083/) (2022)

## Built on by (most cited first, 7 of 7)

- [Compact Latent Manifold Translation: A Parameter-Efficient Foundation Model for Cross-Modal and Cross-Frequency Physiological Signal Synthesis](https://mlsanity.com/paper/2605.13248/) (2026)
- [PRISM-CTG: A Foundation Model for Cardiotocography Analysis with Multi-View SSL](https://mlsanity.com/paper/2605.02917/) (2026)
- [CogAdapt: Adapting Clinical ECG Foundation Models for Wearable Cognitive Load Assessment](https://mlsanity.com/paper/2605.22774/) (2026)
- [ADAPTOOD: Uncertainty-Aware Fine-Tuning for Out-of-Distribution ECG Time Series Models](https://mlsanity.com/paper/2606.04164/) (2026)
- [FOUND-AF: Benchmarking ECG Foundation Models for Atrial Fibrillation Detection](https://mlsanity.com/paper/2608.03597/) (2026)
- [LAEF: A Lead-Agnostic ECG Foundation Model Towards Point-of-Care Diagnostics](https://mlsanity.com/paper/2608.03690/) (2026)
- [PhysioTRACE: Provenance-Aware Stress Tests for Physiological Foundation Models](https://mlsanity.com/paper/2609.34466/) (2026)

## Abstract (arXiv)

The electrocardiogram (ECG) is a key diagnostic tool in cardiovascular health. Single-lead ECG recording is integrated into both clinical-grade and consumer wearables. While self-supervised pretraining of foundation models on unlabeled ECGs improves diagnostic performance, existing approaches do not incorporate domain knowledge from clinical metadata. We introduce a novel contrastive learning approach that utilizes an established clinical risk score to adaptively weight negative pairs: clinically-guided contrastive learning. It aligns the similarities of ECG embeddings with clinically meaningful differences between subjects, with an explicit mechanism to handle missing metadata. On 12-lead ECGs from 161K patients in the MIMIC-IV dataset, we pretrain single-lead ECG foundation models at three scales, collectively called CLEF, using only routinely collected metadata without requiring per-sample ECG annotations. We evaluate CLEF on 18 clinical classification and regression tasks across 7 held-out datasets, and benchmark against 5 foundation model baselines and 3 self-supervised algorithms. When pretrained on 12-lead ECG data and tested on lead-I data, CLEF outperforms self-supervised foundation model baselines: the medium-sized CLEF achieves average AUROC improvements of at least 2.6% in classification and average reductions in MAEs of at least 3.2% in regression. Comparing with existing self-supervised learning algorithms, CLEF improves the average AUROC by at least 1.8%. Moreover, when pretrained only on lead-I data for classification tasks, CLEF performs comparably to the state-of-the-art ECGFounder, which was trained in a supervised manner. Overall, CLEF enables more accurate and scalable single-lead ECG analysis, advancing remote health monitoring. Code and pretrained CLEF models are available at: github.com/Nokia-Bell-Labs/ecg-foundation-model.