English(EN) Beyond Patient Invariance: Learning Cardiac Dynamics via Action-Conditioned JEPAs

超越患者不变性：通过动作条件化 JEPAs 学习心脏动力学

作者 PulseAugur 编辑部 · [2 个来源] · 2026-04-24 14:47

研究人员开发了一种新的医疗保健领域自监督学习方法，超越了传统的不变性方法。他们提出的动作条件化世界模型旨在通过预测未来的生理状态来模拟疾病进展，从而将稳定的解剖学特征与动态的病理变化分离开来。该方法改编自 LeJEPA 框架，并在 MIMIC-IV-ECG 数据集上进行了评估，在关键的分诊任务中表现优于完全监督基线，并在资源匮乏的情况下显示出更高的样本效率。 AI

影响这项研究可能为临床诊断带来更具样本效率和鲁棒性的 AI 模型，尤其是在数据稀缺的环境中。

排序理由学术论文，介绍了一种用于医疗保健领域自监督学习的新颖方法。

在 arXiv cs.LG 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。我们如何撰写摘要 →

报道来源 [2]

arXiv cs.LG TIER_1 English(EN) · Jose Geraldo Fernandes, Luiz Facury, Pedro Robles Dutenhefner, Wagner Meira Jr · 2026-04-27 04:00

Beyond Patient Invariance: Learning Cardiac Dynamics via Action-Conditioned JEPAs

arXiv:2604.22618v1 Announce Type: new Abstract: 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 misa…
arXiv cs.LG TIER_1 English(EN) · Wagner Meira · 2026-04-24 14:47

Beyond Patient Invariance: Learning Cardiac Dynamics via Action-Conditioned JEPAs

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 mathematic…

报道来源 [2]

Beyond Patient Invariance: Learning Cardiac Dynamics via Action-Conditioned JEPAs

Beyond Patient Invariance: Learning Cardiac Dynamics via Action-Conditioned JEPAs

相关话题