Department of Physiology, Cardiovascular Research Institute Maastricht (CARIM), Maastricht, The Netherlands; Bakken Research Center, Medtronic, plc, Maastricht, The Netherlands.
Department of Physiology, Cardiovascular Research Institute Maastricht (CARIM), Maastricht, The Netherlands.
Heart Rhythm. 2023 Sep;20(9):1316-1324. doi: 10.1016/j.hrthm.2023.05.025. Epub 2023 May 27.
Continuous optimization of atrioventricular (AV) delay for cardiac resynchronization therapy (CRT) is mainly performed by electrical means.
The purpose of this study was to develop an estimation model of cardiac function that uses a piezoelectric microphone embedded in a pulse generator to guide CRT optimization.
Electrocardiogram, left ventricular pressure (LVP), and heart sounds were simultaneously collected during CRT device implantation procedures. A piezoelectric alarm transducer embedded in a modified CRT device facilitated recording of heart sounds in patients undergoing a pacing protocol with different AV delays. Machine learning (ML) was used to produce a decision-tree ensemble model capable of estimating absolute maximal LVP (LVP) and maximal rise of LVP (LVdP/dt) using 3 heart sound-based features. To gauge the applicability of ML in AV delay optimization, polynomial curves were fitted to measured and estimated values.
In the data set of ∼30,000 heartbeats, ML indicated S1 amplitude, S2 amplitude, and S1 integral (S1 energy for LVdP/dt) as most prominent features for AV delay optimization. ML resulted in single-beat estimation precision for absolute values of LVP and LVdP/dt of 67% and 64%, respectively. For 20-30 beat averages, cross-correlation between measured and estimated LVP and LVdP/dt was 0.999 for both. The estimated optimal AV delays were not significantly different from those measured using invasive LVP (difference -5.6 ± 17.1 ms for LVP and +5.1 ± 6.7 ms for LVdP/dt). The difference in function at estimated and measured optimal AV delays was not statiscally significant (1 ± 3 mm Hg for LVP and 9 ± 57 mm Hg/s for LVdP/dt).
Heart sound sensors embedded in a CRT device, powered by a ML algorithm, provide a reliable assessment of optimal AV delays and absolute LVP and LVdP/dt.
心脏再同步治疗（CRT）的房室（AV）延迟持续优化主要通过电手段进行。
本研究旨在开发一种使用嵌入脉冲发生器中的压电麦克风来指导 CRT 优化的心脏功能估计模型。
在 CRT 设备植入过程中同时记录心电图、左心室压力（LVP）和心音。在修改后的 CRT 设备中嵌入压电报警传感器，便于在不同 AV 延迟起搏方案下记录心音。使用机器学习（ML）生成决策树集成模型，该模型能够使用 3 个基于心音的特征来估计绝对最大 LVP（LVP）和最大 LVP 上升（LVdP/dt）。为了评估 ML 在 AV 延迟优化中的适用性，使用多项式曲线拟合实测值和估计值。
在约 30000 次心跳的数据集上，ML 表明 S1 幅度、S2 幅度和 S1 积分（LVdP/dt 的 S1 能量）是 AV 延迟优化最显著的特征。ML 在心音单拍估计中，LVP 和 LVdP/dt 的绝对值精度分别为 67%和 64%。对于 20-30 次心跳的平均值，测量值和估计值之间的 LVP 和 LVdP/dt 的互相关系数均为 0.999。估计的最佳 AV 延迟与使用侵入性 LVP 测量的最佳 AV 延迟没有显著差异（LVP 差异为-5.6±17.1 ms，LVdP/dt 差异为+5.1±6.7 ms）。在估计和测量的最佳 AV 延迟下的功能差异无统计学意义（LVP 为 1±3 mmHg，LVdP/dt 为 9±57 mmHg/s）。
嵌入 CRT 设备的心脏传感器，通过 ML 算法提供可靠的最佳 AV 延迟和绝对 LVP 和 LVdP/dt 评估。
