Optical Mapping-Validated Machine Learning Improves Atrial Fibrillation Driver Detection by Multi-Electrode Mapping.
Optical Mapping-Validated Machine Learning Improves Atrial Fibrillation Driver Detection by Multi-Electrode Mapping.
复制标题
DOI：
10.1161/circep.119.008249
复制
发表时间：
2020-10
期刊：
影响因子：
--
通讯作者：
Fedorov VV
中科院分区：
文献类型：
--
作者：
Zolotarev AM;Hansen BJ;Ivanova EA;Helfrich KM;Li N;Janssen PML;Mohler PJ;Mokadam NA;Whitson BA;Fedorov MV;Hummel JD;Dylov DV;Fedorov VV
心房颤动 (AF) 可由局部壁内折返驱动器维持。然而，通过临床仅表面多电极映射 (MEM) 进行的 AF 驱动器检测依赖于激活图的主观解释。我们假设将机器学习 (ML) 应用于电描记图频谱可以通过 MEM 准确地实现驾驶员检测的自动化，并为 MEM 结果的解释增加一些客观性。通过地下近红外光学测绘 (NIOM)（0.3mm2 分辨率）和 64 电极 MEM（分别具有 3mm2 和 9mm2 分辨率的高密度 (HD) 或低密度 (LD)），在移植的人类心房 (n=11) 中同时映射时间和空间稳定的单个 AF 驱动器。单极 MEM 和 NIOM 记录通过傅里叶变换分析处理为 28407 个总傅里叶光谱。从每个傅里叶频谱中提取了 35 个 ML 特征。有针对性的驱动器消融和 NIOM 激活图有效地定义了 AF 驱动器优先轨道的中心和外围，并为 MEM 阵列中的驱动器电极与非驱动器电极提供了经过验证的分类。与分析单个电描记图频率特征相比，对每个周围 8 个电极邻域的特征进行平均，显着改善了 AF 驱动电描记图的分类。当将包括驾驶员外围电极在内的不太严格的注释添加到驾驶员中心注释时，分类指标会增加。值得注意的是，HD 导管数据集二元分类的 f1 分数显着高于 LD 导管（0.81 ± 0.02 vs 0.66 ± 0.04，p<0.05）。经过训练的算法正确突出显示了 HD 驱动器区域的 86%，但 LD MEM 阵列仅正确突出显示了 80%（LD+HD 阵列合计为 81%）。与 NIOM 黄金标准相比，根据傅立叶频谱特征预先训练的 ML 模型可以将电描记图记录有效分类为 AF 驾驶员或非驾驶员。 NIOM 验证的 ML 方法的未来应用可能会提高 AF 驱动器检测的准确性，以用于患者的靶向消融治疗。
Atrial fibrillation (AF) can be maintained by localized intramural reentrant drivers. However, AF driver detection by clinical surface-only multi-electrode mapping (MEM) has relied on subjective interpretation of activation maps. We hypothesized that application of Machine Learning (ML) to electrogram frequency spectra may accurately automate driver detection by MEM and add some objectivity to the interpretation of MEM findings. Temporally and spatially stable single AF drivers were mapped simultaneously in explanted human atria (n=11) by subsurface near-infrared optical mapping (NIOM) (0.3mm2 resolution) and 64-electrode MEM (Higher-Density (HD) or Lower-Density (LD) with 3mm2 and 9mm2 resolution, respectively). Unipolar MEM and NIOM recordings were processed by Fourier Transform analysis into 28407 total Fourier spectra. Thirty-five features for ML were extracted from each Fourier spectrum. Targeted driver ablation and NIOM activation maps efficiently defined the center and periphery of AF driver preferential tracks and provided validated classifications for driver vs non-driver electrodes in MEM arrays. Compared to analysis of single electrogram frequency features, averaging the features for each surrounding 8 electrodes neighborhood, significantly improved classification of AF driver electrograms. The classification metrics increased when less strict annotation including driver periphery electrodes were added to driver center annotation. Notably, f1-score for the binary classification of HD catheter dataset were significantly higher than that of LD catheter (0.81 ± 0.02 vs 0.66 ± 0.04, p<0.05). The trained algorithm correctly highlighted 86% of driver regions with HD but only 80% with LD MEM arrays (81% for LD+HD arrays together). The ML model pre-trained on Fourier spectrum features allows efficient classification of electrograms recordings as AF driver or non-driver compared to the NIOM gold-standard. Future application of NIOM-validated ML approach may improve the accuracy of AF driver detection for targeted ablation treatment in patients.