用自监督学习从少量超声视频预测心输出量，效果超过百万例训练的模型
Cardiac Output Prediction from Echocardiograms: Self-Supervised Learning with Limited Data
- 在数据稀缺下用SimCLR做自监督预训练，提升超声视频表征能力
- 测试集皮尔逊相关系数达0.41，优于基于百万例数据训练的PanEcho模型
- 适合医疗影像中数据少、需高效学习的场景，尤其心脏超声分析
心输出量（CO）是心血管疾病诊断与管理的关键指标。然而，其准确测量依赖于右心导管术，该方法侵入性强且耗时，推动了基于超声心动图的非侵入性替代方案的发展。本文提出一种基于SimCLR的自监督学习（SSL）预训练策略，用于从心尖四腔超声视频中预测心输出量。预训练在下游任务相同的有限数据集上进行，展示了在数据稀缺条件下SSL的潜力。结果表明，SSL有效缓解过拟合，改善表示学习，在测试集上平均皮尔逊相关系数达到0.41，优于在超过一百万例超声检查上训练的PanEcho模型。源代码已公开于https://github.com/EIDOSLAB/cardiac-output。
原文摘要 · Abstract (English)
Cardiac Output (CO) is a key parameter in the diagnosis and management of cardiovascular diseases. However, its accurate measurement requires right-heart catheterization, an invasive and time-consuming procedure, motivating the development of reliable non-invasive alternatives using echocardiography. In this work, we propose a self-supervised learning (SSL) pretraining strategy based on SimCLR to improve CO prediction from apical four-chamber echocardiographic videos. The pretraining is performed using the same limited dataset available for the downstream task, demonstrating the potential of SSL even under data scarcity. Our results show that SSL mitigates overfitting and improves representation learning, achieving an average Pearson correlation of 0.41 on the test set and outperforming PanEcho, a model trained on over one million echocardiographic exams. Source code is available at https://github.com/EIDOSLAB/cardiac-output.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。