Ravikumar Vasanth, Tolkacheva Elena G
Annu Int Conf IEEE Eng Med Biol Soc. 2020 Jul;2020:414-417. doi: 10.1109/EMBC44109.2020.9175773.
Atrial Fibrillation (AF) is most common sustained cardiac arrhythmia and a precursor to many fatal cardiac conditions. Catheter ablation, which is a minimally invasive treatment, is associated with limited success rates in patients with persistent AF. Rotors are believed to maintain AF and core of rotors are considered to be robust targets for ablation. Recently, multiscale entropy (MSE) was proposed to identify the core of rotors in ex-vivo rabbit hearts. However, MSE technique is sensitive to intrinsic parameters, such as scale factor and template dimension, that may lead to an imprecise estimation of entropy measures. The purpose of this research is optimize MSE approach to improve its accuracy and sensitivity in rotor core identification using simulated EGMs from human atrial model. Specifically, we have identified the optimal time scale factor (τopt) and optimal template dimension (Τopt) that are needed for efficient rotor core identification. The τopt was identified to be 10, using a convergence graph, and the Τopt (~20 ms) remained the same at different sampling rates, indicating that optimized MSE will be efficient in identifying core of the rotor irrespective of the signal acquisition system.
心房颤动（AF）是最常见的持续性心律失常，也是许多致命心脏疾病的先兆。导管消融是一种微创治疗方法，对于持续性AF患者，其成功率有限。转子被认为维持着AF，而转子的核心被视为消融的有力靶点。最近，多尺度熵（MSE）被提出用于识别离体兔心脏中的转子核心。然而，MSE技术对诸如尺度因子和模板维度等内在参数敏感，这可能导致熵测量的估计不准确。本研究的目的是优化MSE方法，以提高其在使用来自人体心房模型的模拟心内电图（EGM）识别转子核心时的准确性和灵敏度。具体而言，我们确定了有效识别转子核心所需的最佳时间尺度因子（τopt）和最佳模板维度（Τopt）。通过收敛图确定τopt为10，并且Τopt（约20毫秒）在不同采样率下保持不变，这表明优化后的MSE无论信号采集系统如何，都能有效地识别转子核心。
