JACC Research Article Aug 25, 2020: 76 (8), 10.1016/j.jacc.2020.06.061 本文由“天纳”临床学术信息人工智能系统自动翻译Left ventricular (LV) diastolic dysfunction is recognized as playing a major role in the pathophysiology of heart failure; however, clinical tools for identifying diastolic dysfunction before echocardiography remain imprecise.左室舒张功能不全在心力衰竭的病理生理学中起着重要的作用，然而，在超声心动图之前确定舒张功能不全的临床工具仍然不精确。This study sought to develop machine-learning models that quantitatively estimate myocardial relaxation using clinical and electrocardiography (ECG) variables as a first step in the detection of LV diastolic dysfunction.这项研究试图发展一种机器学习模型，利用临床和心电图（ECG）变量作为检测左室舒张功能不全的第一步，定量评估心肌舒张功能。A multicenter prospective study was conducted at 4 institutions in North America enrolling a total of 1,202 subjects. Patients from 3 institutions (n = 814) formed an internal cohort and were randomly divided into training and internal test sets (80:20). Machine-learning models were developed using signal-processed ECG, traditional ECG, and clinical features and were tested using the test set. Data from the fourth institution was reserved as an external test set (n = 388) to evaluate the model generalizability.一项多中心前瞻性研究在北美的4个机构进行，共有1202名受试者。来自3个机构（n=814）的患者组成一个内部队列，随机分为训练组和内部测试组（80:20）。机器学习模型是利用信号处理后的心电图、传统心电图和临床特征开发的，并使用测试集进行测试。来自第四研究所的数据被保留为外部测试集（n=388），以评估模型的可推广性。Despite diversity in subjects, the machine-learning model predicted the quantitative values of the LV relaxation velocities (e’) measured by echocardiography in both internal and external test sets (mean absolute error: 1.46 and 1.93 cm/s; adjusted R2 = 0.57 and 0.46, respectively). Analysis of the area under the receiver operating characteristic curve (AUC) revealed that the estimated eʹ discriminated the guideline-recommended thresholds for abnormal myocardial relaxation and diastolic and systolic dysfunction (LV ejection fraction) the internal (area under the curve [AUC]: 0.83, 0.76, and 0.75) and external test sets (0.84, 0.80, and 0.81), respectively. Moreover, the estimated eʹ allowed prediction of LV diastolic dysfunction based on multiple age- and sex-adjusted reference limits (AUC: 0.88 and 0.94 in the internal and external sets, respectively).尽管受试者存在差异，但机器学习模型预测了通过超声心动图测量的左室舒张速度（e’）的定量值（平均绝对误差：1.46和1.93 cm/s；校正后的R2分别为0.57和0.46）。对受试者操作特性曲线（AUC）下面积的分析表明，估计的eʹ可区分异常心肌舒张和舒张和收缩功能障碍（LV射血分数）的指导建议阈值，内部（曲线下面积[AUC]：0.83，0.76，和0.75）和外部测试集（0.84、0.80和0.81）。此外，估计的eʹ可以根据多个年龄和性别调整后的参考限值预测左室舒张功能不全（内部和外部的AUC分别为0.88和0.94）。A quantitative prediction of myocardial relaxation can be performed using easily obtained clinical and ECG features. This cost-effective strategy may be a valuable first clinical step for assessing the presence of LV dysfunction and may potentially aid in the early diagnosis and management of heart failure patients.利用容易获得的临床和心电图特征，可以进行心肌舒张的定量预测。这种成本效益高的策略可能是评估左室功能不全的临床第一步，可能有助于心力衰竭患者的早期诊断和治疗。
This study was supported in part by funds from the National Science Foundation (NSF: #1920920) and by Heart Test Laboratories, Inc. d/b/a HeartSciences. HeartSciences provided funding and spECG devices. They had no role in developing the research plan, analysis, drafting the manuscript other than providing necessary resources to collect the information from different site investigators. Dr. Kagiyama has been supported by a research grant from Hitachi Healthcare. Dr. Sengupta has served as a consultant to HeartSciences, Ultromics, and Kencor Health. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. John Brush, MD, served as Guest Associate Editor for this paper. Deepak L. Bhatt, MD, MPH, served as Guest Editor-in-Chief for this paper.The authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC .Listen to this manuscript's audio summary by Editor-in-Chief Dr. Valentin Fuster onJACC.org.1. G. Savarese, L.H. Lund. Global public health burden of heart failure. Card Fail Rev2017;3:7-11. doi: 10.15420/cfr.2016:25:22. F.A. Flachskampf, T. Biering-Sorensen, S.D. Solomon, O. Duvernoy, T. Bjerner, O.A. Smiseth. Cardiac imaging to evaluate left ventricular diastolic function. J Am Coll Cardiol Img2015;8:1071-1093. doi: 3. W.A. Aljaroudi, J.D. Thomas, L.L. Rodriguez, W.A. Jaber. Prognostic value of diastolic dysfunction: state of the art review. Cardiol Rev2014;22:79-90. doi: 10.1097/CRD.0b013e31829cf7334. M. Kloch-Badelek, T. Kuznetsova, W. Sakiewicz. Prevalence of left ventricular diastolic dysfunction in European populations based on cross-validated diagnostic thresholds.. Cardiovasc Ultrasound2012;10:10-10. doi: 10.1186/1476-7120-10-105. M. Fischer, A. Baessler, H.W. Hense. Prevalence of left ventricular diastolic dysfunction in the community: Results from a Doppler echocardiographic-based survey of a population sample. Eur Heart J2003;24:320-328. doi: 10.1016/S0195-668X(02)00428-16. Z.I. Attia, S. Kapa, F. Lopez-Jimenez. Screening for cardiac contractile dysfunction using an artificial intelligence-enabled electrocardiogram. Nat Med2019;25:70-74. doi: 7. P.P. Sengupta, H. Kulkarni, J. Narula. Prediction of Abnormal Myocardial Relaxation From Signal Processed Surface ECG. JACC2018;71:1650-1660. doi: 10.1016/j.jacc.2018.02.0248. N.S. Chahal, T.K. Lim, P. Jain, J.C. Chambers, J.S. Kooner, R. Senior. Normative reference values for the tissue Doppler imaging parameters of left ventricular function: a population-based study. Eur J Echocardiogr2010;11:51-56. doi: 10.1093/ejechocard/jep1649. H. Dalen, A. Thorstensen, L.J. Vatten, S.A. Aase, A. Stoylen. Reference values and distribution of conventional echocardiographic Doppler measures and longitudinal tissue Doppler velocities in a population free from cardiovascular disease. Circ Cardiovasc Imaging2010;3:. doi: 10.1161/CIRCIMAGING.109.92602210. M. Nayor, L.L. Cooper, D.M. Enserro. Left ventricular diastolic dysfunction in the community: impact of diagnostic criteria on the burden, correlates, and prognosis. J Am Heart Assoc2018;7:. doi: 11. J.K. Oh, W.R. Miranda, J.G. Bird, G.C. Kane, S.F. Nagueh. The 2016 diastolic function guideline: is it already time to revisit or revise them?. J Am Coll Cardiol Img2020;13:327-335. doi: 12. H. Okura, Y. Takada, A. Yamabe. Age- and Gender-Specific Changes in the Left Ventricular Relaxation: A Doppler Echocardiographic Study in Healthy Individuals. Circ Cardiovasc Imaging2009;2:41-46. doi: 10.1161/CIRCIMAGING.108.80908713. R.M. Lang, L.P. Badano, V. Mor-Avi. Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echocardiogr2015;28:1-39.e14. doi: 10.1016/j.echo.2014.10.00314. N. Kagiyama, S. Shrestha, P.D. Farjo, P.P. Sengupta. Artificial intelligence: practical primer for clinical research in cardiovascular disease. J Am Heart Assoc2019;8:. doi: 15. J.A. Crowe, N.M. Gibson, M.S. Woolfson, M.G. Somekh. Wavelet transform as a potential tool for ECG analysis and compression.. J Biomed Eng1992;14:268-272. doi: 10.1016/0141-5425(92)90063-Q16. M. Engin, O. Cidam, E.Z. Engin. Wavelet transformation based watermarking technique for human electrocardiogram (ECG).. J Med Syst2005;29:589-594. doi: 17. S.F. Nagueh, O.A. Smiseth, C.P. Appleton. Recommendations for the evaluation of left ventricular diastolic function by echocardiography: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echocardiogr2016;29:277-314. doi: 10.1016/j.echo.2016.01.01118. M.B. Kursa, W.R. Rundnicki. Feature Selection with the Boruta Package. J Stat Softw2010;36:1-13. doi: 10.18637/jss.v036.i1119. L. Breiman. Random Forests. Machine Learning2001;45:5-32. doi: 10.1017/CBO9781107415324.00420. . Walking the tightrope of artificial intelligence guidelines in clinical practice. The Lancet Digital Health2019;1:. doi: 21. W. Aljaroudi, M.C. Alraies, C. Halley. Impact of progression of diastolic dysfunction on mortality in patients with normal ejection fraction. Circulation2012;125:. doi: 10.1161/CIRCULATIONAHA.111.06642322. K. Wachtell, V. Palmieri, E. Gerdts. Prognostic significance of left ventricular diastolic dysfunction in patients with left ventricular hypertrophy and systemic hypertension (the LIFE Study).. Am J Cardiol2010;106:999-1005. doi: 10.1016/j.amjcard.2010.05.03223. S.D. Solomon, A. Verma, A. Desai. Effect of intensive versus standard blood pressure lowering on diastolic function in patients with uncontrolled hypertension and diastolic dysfunction. Hypertension2010;55:241-248. doi: 10.1161/HYPERTENSIONAHA.109.13852924. H. Takagi, Y. Mizuno, M. Niwa, S.N. Goto, T. Umemoto. A meta-analysis of randomized controlled trials of azilsartan therapy for blood pressure reduction. Hypertens Res2014;37:432-437. doi: 25. M. Leung, V.W. Wong, M. Hudson, D.Y. Leung. Impact of improved glycemic control on cardiac function in type 2 diabetes mellitus. Circ Cardiovasc Imaging2016;9:. doi: 26. . Health, United States, 2017: With Special Feature on Mortality. 2018;:. doi: 27. R.E. Carter, Z.I. Attia, F. Lopez-Jimenez, P.A. Friedman. Pragmatic considerations for fostering reproducible research in artificial intelligence. NPJ Digit Med2019;2:42. doi: 28. A.E. Ivanescu, P. Li, B. George. The importance of prediction model validation and assessment in obesity and nutrition research. Int J Obes (Lond)2016;40:887-894. doi: 10.1038/ijo.2015.21429. D.A. Kass, J.G. Bronzwaer, W.J. Paulus. What Mechanisms Underlie Diastolic Dysfunction in Heart Failure?. Circ. Res.2004;94:1533-1542. doi: 10.1161/01.RES.0000129254.25507.d630. M.R. Zile, C.F. Baicu, W.H. Gaasch. Diastolic Heart Failure -- Abnormalities in Active Relaxation and Passive Stiffness of the Left Ventricle. NEJM2004;350:1953-1959. doi: 10.1056/NEJMoa032566
JACC Research Article Aug 25, 2020: 76 (8), 10.1016/j.jacc.2020.06.061