从稀缺标签中学习：多视图超声心动图用于射血分数预测
Learning from Scarce Labels: Multi-View Echocardiography for Ejection Fraction Prediction
浏览论文内容
中文总结 AI 辅助
该研究构建含25000+PLAX视频的标注数据集，训练出MAE6.86%的PLAX EF模型，经多视图融合后MAE达6.37%，相关资源已公开，为心尖视图不可行场景提供临床可行的EF估计方案
中文摘要 AI 辅助
据我们所知，我们推出了首个公开可用的、用于从胸骨旁长轴（PLAX）超声心动图预测左心室射血分数（EF）的资源。由于此前不存在PLAX-EF数据集，我们的工作聚焦于一种创新的数据生成策略以克服数据稀缺问题。通过利用临床记录与超声心动图视频之间的时间相关性，结合微调视图分类器与代理标注，我们构建了一个包含超过25000个PLAX视频的标注数据集。这使我们能够训练首个可复现的PLAX EF模型，其平均绝对误差（MAE）为6.86%。作为临床标准的心尖四腔（A4C）方法报告的MAE值为6%-7%，我们的结果表明，从PLAX视图估计EF是可行的且具有临床相关性。这超过了现有方法的性能，并为心尖视图可能不可行的情况提供了具有临床相关性的解决方案。进一步，我们证明通过简单的无权重后期融合结合PLAX和A4C预测，可将单视图基线的MAE提升至6.37%，凸显了多视图整合的价值。为推动后续研究，我们在GitHub、Hugging Face和Google Colab上发布了数据集标注、训练好的模型和可运行演示：该https URL
英文摘要
We present, to the best of our knowledge, the first publicly available resource for predicting left ventricular ejection fraction (EF) from parasternal long-axis (PLAX) echocardiography. Because no PLAX-EF datasets previously existed, our work focuses on an innovative data generation strategy to overcome this scarcity. By leveraging a time-based correlation between clinical notes and echocardiographic videos, combined with fine-tuning view classifiers and proxy labeling, we created a labeled dataset of over 25,000 PLAX videos. This enables us to train the first reproducible PLAX EF model, achieving a mean absolute error (MAE) of 6.86%. Given that apical four-chamber (A4C) methods, the clinical standard, report MAE values of 6%-7%, our results demonstrate that EF estimation from PLAX views is both feasible and clinically relevant. This surpasses the performance of existing methods and provides a clinically relevant solution for situations where apical views may not be feasible. Going further, we demonstrate that combining PLAX and A4C predictions via simple unweighted late fusion improves both single-view baselines to a 6.37% MAE, underscoring the value of multi-view integration. To promote continued research, we release the dataset labels, trained models, and runnable demos on GitHub, Hugging Face, and Google Colab: https://github.com/Jeffrey4899/PLAX_EF_Labels_202509
发表机构
- California Institute of Technology（加州理工学院）
- Asari AI
机构由 AI 辅助整理，请以论文原文为准。
