MedStar Health Research Institute and Georgetown University, Washington, District of Columbia.
Ultromics, Oxford, United Kingdom.
J Am Soc Echocardiogr. 2022 Dec;35(12):1226-1237.e7. doi: 10.1016/j.echo.2022.07.004. Epub 2022 Jul 19.
Transthoracic echocardiography is the leading cardiac imaging modality for patients admitted with COVID-19, a condition of high short-term mortality. The aim of this study was to test the hypothesis that artificial intelligence (AI)-based analysis of echocardiographic images could predict mortality more accurately than conventional analysis by a human expert.
Patients admitted to 13 hospitals for acute COVID-19 who underwent transthoracic echocardiography were included. Left ventricular ejection fraction (LVEF) and left ventricular longitudinal strain (LVLS) were obtained manually by multiple expert readers and by automated AI software. The ability of the manual and AI analyses to predict all-cause mortality was compared.
In total, 870 patients were enrolled. The mortality rate was 27.4% after a mean follow-up period of 230 ± 115 days. AI analysis had lower variability than manual analysis for both LVEF (P = .003) and LVLS (P = .005). AI-derived LVEF and LVLS were predictors of mortality in univariable and multivariable regression analysis (odds ratio, 0.974 [95% CI, 0.956-0.991; P = .003] for LVEF; odds ratio, 1.060 [95% CI, 1.019-1.105; P = .004] for LVLS), but LVEF and LVLS obtained by manual analysis were not. Direct comparison of the predictive value of AI versus manual measurements of LVEF and LVLS showed that AI was significantly better (P = .005 and P = .003, respectively). In addition, AI-derived LVEF and LVLS had more significant and stronger correlations to other objective biomarkers of acute disease than manual reads.
AI-based analysis of LVEF and LVLS had similar feasibility as manual analysis, minimized variability, and consequently increased the statistical power to predict mortality. AI-based, but not manual, analyses were a significant predictor of in-hospital and follow-up mortality.
经胸超声心动图是 COVID-19 患者入院后的主要心脏成像方式，COVID-19 患者短期死亡率较高。本研究旨在验证一个假设，即基于人工智能（AI）的超声心动图图像分析比人类专家的常规分析更能准确预测死亡率。
纳入了 13 家医院因急性 COVID-19 入院并接受经胸超声心动图检查的患者。左心室射血分数（LVEF）和左心室纵向应变（LVLS）由多名专家读者和自动 AI 软件手动获得。比较手动和 AI 分析预测全因死亡率的能力。
共纳入 870 例患者，平均随访 230±115 天后死亡率为 27.4%。与手动分析相比，AI 分析在 LVEF（P=0.003）和 LVLS（P=0.005）方面的变异性均较低。AI 衍生的 LVEF 和 LVLS 在单变量和多变量回归分析中是死亡率的预测因素（LVEF 的优势比，0.974[95%置信区间，0.956-0.991；P=0.003]；LVLS 的优势比，1.060[95%置信区间，1.019-1.105；P=0.004]），而手动分析的 LVEF 和 LVLS 则不是。直接比较 AI 与手动测量 LVEF 和 LVLS 的预测价值表明，AI 明显更好（P=0.005 和 P=0.003）。此外，与其他急性疾病的客观生物标志物相比，AI 衍生的 LVEF 和 LVLS 具有更显著和更强的相关性。
LVEF 和 LVLS 的 AI 分析与手动分析具有相似的可行性，最小化了变异性，从而提高了预测死亡率的统计能力。基于 AI 的分析，而不是手动分析，是院内和随访死亡率的一个显著预测因素。
