基于 12 导联 QRS 波群的机器学习以识别具有不同预后的心脏再同步治疗患者。
Machine Learning of 12-Lead QRS Waveforms to Identify Cardiac Resynchronization Therapy Patients With Differential Outcomes.
机构信息
Cleveland Clinic Lerner College of Medicine (A.K.F., N.V., M.J.N., E.Z.G., R.A.G., J.B., M.K.C.), Case Western Reserve University, Cleveland, OH.
Department of Cardiovascular Medicine, Heart and Vascular Institute (J.R., K.M.T., D.P., S.T., L.A.M., N.V., M.J.N., E.Z.G.), Cleveland Clinic, OH.
出版信息
Circ Arrhythm Electrophysiol. 2020 Jul;13(7):e008210. doi: 10.1161/CIRCEP.119.008210. Epub 2020 Jun 14.
BACKGROUND
Cardiac resynchronization therapy (CRT) improves heart failure outcomes but has significant nonresponse rates, highlighting limitations in ECG selection criteria: QRS duration (QRSd) ≥150 ms and subjective labeling of left bundle branch block (LBBB). We explored unsupervised machine learning of ECG waveforms to identify CRT subgroups that may differentiate outcomes beyond QRSd and LBBB.
METHODS
We retrospectively analyzed 946 CRT patients with conduction delay. Principal component analysis (PCA) dimensionality reduction obtained a 2-dimensional representation of preCRT 12-lead QRS waveforms. -means clustering of the 2-dimensional PCA representation of 12-lead QRS waveforms identified 2 patient subgroups (QRS PCA groups). Vectorcardiographic QRS area was also calculated. We examined following 2 primary outcomes: (1) composite end point of death, left ventricular assist device, or heart transplant, and (2) degree of echocardiographic left ventricular ejection fraction (LVEF) change after CRT.
RESULTS
Compared with QRS PCA Group 2 (=425), Group 1 (=521) had lower risk for reaching the composite end point (HR, 0.44 [95% CI, 0.38-0.53]; <0.001) and experienced greater mean LVEF improvement (11.1±11.7% versus 4.8±9.7%; <0.001), even among patients with LBBB with QRSd ≥150 ms (HR, 0.42 [95% CI, 0.30-0.57]; <0.001; mean LVEF change 12.5±11.8% versus 7.3±8.1%; =0.001). QRS area also stratified outcomes but had significant differences from QRS PCA groups. A stratification scheme combining QRS area and QRS PCA group identified patients with LBBB with similar outcomes to non-LBBB patients (HR, 1.32 [95% CI, 0.93-1.62]; difference in mean LVEF change: 0.8% [95% CI, -2.1% to 3.7%]). The stratification scheme also identified patients with LBBB with QRSd <150 ms with comparable outcomes to patients with LBBB with QRSd ≥150 ms (HR, 0.93 [95% CI, 0.67-1.29]; difference in mean LVEF change: -0.2% [95% CI, -2.7% to 3.0%]).
CONCLUSIONS
Unsupervised machine learning of ECG waveforms identified CRT subgroups with relevance beyond LBBB and QRSd. This method may assist in objective classification of bundle branch block morphology in CRT.
背景
心脏再同步治疗（CRT）改善心力衰竭的预后，但仍有显著的无反应率，这突出了心电图选择标准的局限性：QRS 时限（QRSd）≥150ms 和主观标记的左束支传导阻滞（LBBB）。我们探索了心电图波形的无监督机器学习，以确定 CRT 亚组，这些亚组可能会在 QRSd 和 LBBB 之外区分预后。
方法
我们回顾性分析了 946 例有传导延迟的 CRT 患者。主成分分析（PCA）降维得到了预 CRT12 导联 QRS 波群的二维表示。12 导联 QRS 波群的二维 PCA 表示的 -means 聚类确定了 2 个患者亚组（QRS PCA 组）。还计算了心向量心电图的 QRS 区域。我们研究了以下 2 个主要结果：（1）死亡、左心室辅助装置或心脏移植的复合终点，（2）CRT 后超声心动图左心室射血分数（LVEF）变化的程度。
结果
与 QRS PCA 组 2（=425）相比，组 1（=521）达到复合终点的风险较低（HR，0.44[95%CI，0.38-0.53]；<0.001），并且 LVEF 改善程度更高（11.1±11.7%对 4.8±9.7%；<0.001），即使在 QRSd≥150ms 的 LBBB 患者中也是如此（HR，0.42[95%CI，0.30-0.57]；<0.001；平均 LVEF 变化 12.5±11.8%对 7.3±8.1%；=0.001）。QRS 区域也对结果进行了分层，但与 QRS PCA 组有显著差异。一种结合 QRS 区域和 QRS PCA 组的分层方案确定了 LBBB 患者的结果与非 LBBB 患者相似（HR，1.32[95%CI，0.93-1.62]；平均 LVEF 变化差异：0.8%[95%CI，-2.1%至 3.7%]）。该分层方案还确定了 QRSd<150ms 的 LBBB 患者与 QRSd≥150ms 的 LBBB 患者具有可比的结果（HR，0.93[95%CI，0.67-1.29]；平均 LVEF 变化差异：-0.2%[95%CI，-2.7%至 3.0%]）。
结论
心电图波形的无监督机器学习确定了 CRT 亚组，其相关性超出了 LBBB 和 QRSd。该方法可能有助于在 CRT 中对束支传导阻滞形态进行客观分类。
Zotero 插件