发表机构
University of Sherbrooke; INSA; Université Lyon 1; CREATIS; Institut Universitaire de France (IUF); NTNU; St. Olavs Hospital; SINTEF Digital; Levanger Hospital; Nord-Trøndelag Hospital Trust; Hôpital Croix-Rousse; Hospices Civils de Lyon; Hôpital Lyon Sud(舍布鲁克大学; 法国国立应用科学学院; 里昂第一大学; CREATIS实验室; 法国大学研究院(IUF); 挪威科技大学; 圣奥拉夫医院; SINTEF数字研究所; 莱旺厄尔医院; 北特伦德拉格医院信托; 克鲁瓦鲁斯医院; 里昂民用医院集团; 里昂南部医院)
机构由 AI 辅助整理,请以论文原文为准。AI 中文总结
针对深度学习超声心肌追踪中的时域漂移问题,提出扩展TAS-Net并加入持久记忆令牌及教师-学生微调策略,以减少应变漂移、提高临床一致性。
AI 中文摘要
超声心动图的心肌应变是心脏功能的关键生物标志物。最近的深度学习方法在心肌运动追踪方面表现出强大的性能,但往往缺乏生理约束,导致跨心动周期的时域漂移。因此,追踪点可能无法在每个心动周期结束时回到其相对的初始位置,从而产生不准确的应变估计,在某些情况下甚至出现发散。我们提出了一种深度学习框架,用于在心肌追踪过程中补偿漂移。我们扩展了一种最先进的超声心动图追踪方法(TAS-Net),加入了持久记忆令牌,这些令牌在完整心动周期的滑动窗口之间共享信息。随后,在真实超声心动图数据上采用教师-学生微调策略,强制实现生理上一致的周期性运动,同时保持追踪精度。实验表明,全局和区域应变漂移减少,与临床参考的一致性提高,测试-重测可重复性更好,支持在临床实践中更可靠的心肌应变估计。
英文摘要
Myocardial strain from echocardiography is a key biomarker for cardiac function. Recent deep learning methods show strong performance for myocardial motion tracking but often lack physiological constraints, leading to temporal drift across the cardiac cycle. Consequently, tracked points may not return to their relative initial positions at the end of each cardiac cycle, producing inaccurate strain estimates and even divergence in some cases. We propose a deep learning framework that compensates for drift during myocardial tracking. We extend a state-of-the-art echocardiographic tracking method (TAS-Net) with persistent memory tokens that share information across sliding windows over full cardiac cycles. A teacher-student fine-tuning strategy on real echocardiographic data then enforces physiologically consistent cyclic motion while preserving tracking accuracy. Experiments show reduced global and regional strain drift, improved agreement with clinical references, and better test-retest reproducibility, supporting more reliable myocardial strain estimation in clinical practice.
CommentsSTACOM 2026, 10 pages