CardioState-JEPA: Delay-Aware Cross-Modal Learning of a Shared Cardiac Representation

• URL: http://arxiv.org/abs/2608.12944v1
• Date: Thu, 13 Aug 2026 08:21:06 GMT
• Title: CardioState-JEPA: Delay-Aware Cross-Modal Learning of a Shared Cardiac Representation
• Abstract summary: We introduce CardioState-JEPA, a cardiac foundation model to learn a single shared representation jointly across ECG, PPG, and PCG.<n>The model maps heterogeneous waveforms into a common token space, processes them with a single shared Transformer encoder, and learns by predicting masked latent cardiac states.<n>Our encoder improves average PPG classification by 8.2 AUROC points, PCG murmur detection by 18.8 AUROC points, and ECG classification by 15.5 AUROC points over the best self-supervised signal baseline.
• Score: 33.42512865785489
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Electrocardiography (ECG), photoplethysmography (PPG), and phonocardiography (PCG) provide complementary views of the same cardiac cycle, yet existing cardiac foundation models are trained for a single sensing modality, leaving the shared physiology across sensors unexploited. We introduce CardioState-JEPA, a cardiac foundation model to learn a single shared representation jointly across ECG, PPG, and PCG, built on a physiology-aware joint-embedding predictive architecture. The model maps heterogeneous waveforms into a common token space, processes them with a single shared Transformer encoder, and learns by predicting masked latent cardiac states, placing the pretraining target on shared physiology rather than sensor-specific waveform appearance. To handle the temporal offsets between electrical, mechanical, and hemodynamic events, cross-modal prediction uses a learned delay aligner that matches signals at the corresponding cardiac time. Because synchronized multi-sensor recordings are scarce, CardioState-JEPA first learns within-modality structure from abundant unimodal data and then uses paired data to align modalities in latent cardiac time. Evaluated as a frozen encoder across 25 downstream tasks spanning ECG, PPG, and PCG, our encoder improves average PPG classification by 8.2 AUROC points, PCG murmur detection by 18.8 AUROC points, and ECG classification by 15.5 AUROC points over the best self-supervised signal baseline and matches or exceeds cardiac models trained with privileged clinical text or supervised labels on several ECG benchmarks. These results establish that heterogeneous cardiac signals can mutually supervise a single foundation model of cardiac physiology.

Related papers

• ECG-LLM: Foundation Model for ECG-Based Cardiac Reasoning [50.82313761980655]
  ECG-LLM is an ECG-conditioned large language model trained across four cohorts 679,112 ECG studies from 186,409 patients.<n>It is trained on clinically structured question-answer pairs derived from ECG signals, clinical context, CMR, and ECHO.<n>It successfully recovers conventional ECG measurements, such as heart rate, and strongly predicts complex CMR-derived phenotypes.
  arXiv  Detail & Related papers  (2026-07-15T11:06:35Z)
• Learning Cardiac Latent Representations in Vectorcardiogram Space [54.12639543496968]
  Existing methods operate almost exclusively in the observable ECG signal space.<n>We propose learning a unified latent representation of cardiac electrical activity directly in the VCG space.<n>We introduce LVCG, the first general self-supervised representation learning framework designed to operate in this physically grounded latent space.
  arXiv  Detail & Related papers  (2026-05-29T12:48:12Z)
• PG-LRF: Physiology-Guided Latent Rectified Flow for Electro-Hemodynamic PPG-to-ECG Generation [51.79176059799321]
  Photoplethysmography is ubiquitous in wearables but lacks ECG-specific diagnostic morphology.<n>Existing methods rely on statistical alignment and data-driven generation.<n>We propose PG-LRF, a physiology-guided latent flow framework.
  arXiv  Detail & Related papers  (2026-05-09T07:39:58Z)
• Echo2ECG: Enhancing ECG Representations with Cardiac Morphology from Multi-View Echos [53.368813255127115]
  Existing self-supervised methods suffer from a representational mismatch by aligning ECGs to single-view Echos.<n>We propose Echo2ECG, a multimodal self-supervised learning framework that enriches ECG representations with the heart's morphological structure captured in multi-view Echos.<n>We evaluate Echo2ECG as an ECG feature extractor on two clinically relevant tasks.
  arXiv  Detail & Related papers  (2026-03-09T15:39:57Z)
• Simulator and Experience Enhanced Diffusion Model for Comprehensive ECG Generation [52.19347532840774]
  We propose SE-Diff, a novel physiological simulator and experience enhanced diffusion model for ECG generation.<n> SE-Diff integrates a lightweight ordinary differential equation (ODE)-based ECG simulator into the diffusion process via a beat decoder.<n>Extensive experiments on real-world ECG datasets demonstrate that SE-Diff improves both signal fidelity and text-ECG semantic alignment.
  arXiv  Detail & Related papers  (2025-11-13T02:57:10Z)
• AnyECG: Foundational Models for Multitask Cardiac Analysis in Real-World Settings [34.078819572852446]
  Electrocardiogram (ECG) is highly sensitive in detecting acute heart attacks.<n>This paper introduces AnyECG, a foundational model designed to extract robust representations from any real-world ECG data.
  arXiv  Detail & Related papers  (2024-11-17T17:32:58Z)
• Deciphering Heartbeat Signatures: A Vision Transformer Approach to Explainable Atrial Fibrillation Detection from ECG Signals [4.056982620027252]
  We develop a vision transformer approach to identify atrial fibrillation based on single-lead ECG data. A residual network (ResNet) approach is also developed for comparison with the vision transformer approach.
  arXiv  Detail & Related papers  (2024-02-12T11:04:08Z)
• Two-stream Network for ECG Signal Classification [3.222802562733787]
  This paper explores an effective algorithm for automatic classifications of multi-classes of heartbeat types based on ECG. A two-stream architecture is used in this paper and presents an enhanced version of ECG recognition based on this. Results on the MIT-BIH Arrhythmia Database demonstrate that the proposed algorithm performs an accuracy of 99.38%.
  arXiv  Detail & Related papers  (2022-10-05T08:14:51Z)
• Generalizing electrocardiogram delineation: training convolutional neural networks with synthetic data augmentation [63.51064808536065]
  Existing databases for ECG delineation are small, being insufficient in size and in the array of pathological conditions they represent. This article delves has two main contributions. First, a pseudo-synthetic data generation algorithm was developed, based in probabilistically composing ECG traces given "pools" of fundamental segments, as cropped from the original databases, and a set of rules for their arrangement into coherent synthetic traces. Second, two novel segmentation-based loss functions have been developed, which attempt at enforcing the prediction of an exact number of independent structures and at producing closer segmentation boundaries by focusing on a reduced number of samples.
  arXiv  Detail & Related papers  (2021-11-25T10:11:41Z)
• CardioGAN: Attentive Generative Adversarial Network with Dual Discriminators for Synthesis of ECG from PPG [25.305949034527202]
  Electrocardiogram (ECG) is the electrical measurement of cardiac activity. Photoplethysmogram ( PPG) is the optical measurement of changes in blood circulation. We propose CardioGAN, an adversarial model which takes PPG as input and generates ECG as output.
  arXiv  Detail & Related papers  (2020-09-30T20:49:30Z)
• Classification of Arrhythmia by Using Deep Learning with 2-D ECG Spectral Image Representation [3.3426603061273994]
  We propose a two-dimensional (2-D) convolutional neural network (CNN) model for the classification of ECG signals into eight classes. We achieved a state-of-the-art average classification accuracy of 99.11%, which is better than those of recently reported results in classifying similar types of arrhythmias.
  arXiv  Detail & Related papers  (2020-05-14T12:11:41Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.