{"slug": "learning-cardiac-dynamics-via-action-conditioned-jepas", "title": "Learning cardiac dynamics via action-conditioned JEPAs", "summary": "A team of researchers submitted an arXiv paper on 24 April 2026 proposing Action-Conditioned World Models, adapting the LeJEPA framework to physiological time-series to model cardiac disease progression as a transition vector on a patient's latent state rather than a static label. Evaluated on the MIMIC-IV-ECG dataset, the approach outperformed fully supervised baselines on the critical triage task and, in low-resource regimes, beat supervised learning by over 0.05 AUROC. Source code is available at the linked GitHub repository.", "body_md": "# Computer Science > Machine Learning\n\n  [Submitted on 24 Apr 2026]\n\n# Title:Beyond Patient Invariance: Learning Cardiac Dynamics via Action-Conditioned JEPAs\n\n[View PDF](https://arxiv.org/pdf/2604.22618)\n\n[HTML (experimental)](https://arxiv.org/html/2604.22618v1)\n\nAbstract:Self-supervised learning in healthcare has largely relied on invariance-based objectives, which maximize similarity between different views of the same patient. While effective for static anatomy, this paradigm is fundamentally misaligned with clinical diagnosis, as it mathematically compels the model to suppress the transient pathological changes it is intended to detect. We propose a shift towards Action-Conditioned World Models that learn to simulate the dynamics of disease progression, or Event-Conditioned. Adapting the LeJEPA framework to physiological time-series, we define pathology not as a static label, but as a transition vector acting on a patient's latent state. By predicting the future electrophysiological state of the heart given a disease onset, our model explicitly disentangles stable anatomical features from dynamic pathological forces. Evaluated on the MIMIC-IV-ECG dataset, our approach outperforms fully supervised baselines on the critical triage task. Crucially, we demonstrate superior sample efficiency: in low-resource regimes, our world model outperforms supervised learning by over 0.05 AUROC. These results suggest that modeling biological dynamics provides a dense supervision signal that is far more robust than static classification. Source code is available at [this https URL](https://github.com/cljosegfer/lesaude-dynamics)\n\n## Submission history\n\nFrom: Luiz Facury De Souza [\n[view email](https://arxiv.org/show-email/56993524/2604.22618)]\n\n**[v1]** Fri, 24 Apr 2026 14:47:52 UTC (679 KB)\n\n### References & Citations\n\nLoading...\n\n# Bibliographic and Citation Tools\n\nBibliographic Explorer \n\n*(*[What is the Explorer?](https://info.arxiv.org/labs/showcase.html#arxiv-bibliographic-explorer))\nConnected Papers \n\n*(*[What is Connected Papers?](https://www.connectedpapers.com/about))\nLitmaps \n\n*(*[What is Litmaps?](https://www.litmaps.co/))\nscite Smart Citations \n\n*(*[What are Smart Citations?](https://www.scite.ai/))\n# Code, Data and Media Associated with this Article\n\nalphaXiv \n\n*(*[What is alphaXiv?](https://alphaxiv.org/))\nCatalyzeX Code Finder for Papers \n\n*(*[What is CatalyzeX?](https://www.catalyzex.com))\nDagsHub \n\n*(*[What is DagsHub?](https://dagshub.com/))\nGotit.pub \n\n*(*[What is GotitPub?](http://gotit.pub/faq))\nHugging Face \n\n*(*[What is Huggingface?](https://huggingface.co/huggingface))\nScienceCast \n\n*(*[What is ScienceCast?](https://sciencecast.org/welcome))\n# Demos\n\n# Recommenders and Search Tools\n\nInfluence Flower \n\n*(*[What are Influence Flowers?](https://influencemap.cmlab.dev/))\nCORE Recommender \n\n*(*[What is CORE?](https://core.ac.uk/services/recommender))\nIArxiv Recommender\n\n*(*[What is IArxiv?](https://iarxiv.org/about))\n# arXivLabs: experimental projects with community collaborators\n\narXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n\nBoth individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.\n\nHave an idea for a project that will add value for arXiv's community? [**Learn more about arXivLabs**](https://info.arxiv.org/labs/index.html).", "url": "https://wpnews.pro/news/learning-cardiac-dynamics-via-action-conditioned-jepas", "canonical_source": "https://arxiv.org/abs/2604.22618", "published_at": "2026-10-07 23:59:29+00:00", "updated_at": "2026-10-08 00:19:17.719060+00:00", "lang": "en", "topics": ["machine-learning", "ai-research", "artificial-intelligence"], "entities": ["MIMIC-IV-ECG", "LeJEPA", "arXiv", "Luiz Facury De Souza"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/learning-cardiac-dynamics-via-action-conditioned-jepas", "markdown": "https://wpnews.pro/news/learning-cardiac-dynamics-via-action-conditioned-jepas.md", "text": "https://wpnews.pro/news/learning-cardiac-dynamics-via-action-conditioned-jepas.txt", "jsonld": "https://wpnews.pro/news/learning-cardiac-dynamics-via-action-conditioned-jepas.jsonld"}}