Naghavi Morteza, Reeves Anthony P, Atlas Kyle C, Zhang Chenyu, Li Dong, Atlas Thomas, Henschke Claudia I, Wong Nathan D, Roy Sion K, Budoff Matthew J, Yankelevitz David F
HeartLung.AI, Houston, Texas, USA.
Department of Electrical and Computer Engineering, Cornell University, New York, USA.
JACC Adv. 2024 Nov 15;3(11):101300. doi: 10.1016/j.jacadv.2024.101300. eCollection 2024 Nov.
AI-CAC provides more actionable information than the Agatston coronary artery calcium (CAC) score. We have recently shown in the MESA (Multi-Ethnic Study of Atherosclerosis) that AI-CAC automated left atrial (LA) volumetry enabled prediction of atrial fibrillation (AF) as early as 1 year.
In this study, the authors evaluated the performance of AI-CAC LA volumetry versus LA measured by human experts using cardiac magnetic resonance imaging (CMRI) for predicting incident AF and stroke and compared them with Cohorts for Heart and Aging Research in Genomic Epidemiology model for atrial fibrillation (CHARGE-AF) risk score, Agatston score, and N-terminal pro b-type natriuretic peptide (NT-proBNP).
We used 15-year outcomes data from 3,552 asymptomatic individuals (52.2% women, age 61.7 ± 10.2 years) who underwent both CAC scans and CMRI in the MESA baseline examination. CMRI LA volume was previously measured by human experts. Data on NT-proBNP, CHARGE-AF risk score, and the Agatston score were obtained from MESA. Discrimination was assessed using the time-dependent area under the curve.
Over 15 years follow-up, 562 cases of AF and 140 cases of stroke accrued. The area under the curve for AI-CAC versus CMRI volumetry for AF (0.802 vs 0.798) and stroke (0.762 vs 0.751) were not significantly different. AI-CAC LA significantly improved the continuous net reclassification index for prediction of 5-year AF when added to CHARGE-AF risk score (0.23), NT-proBNP (0.37, 0.37), and Agatston score (0.44) ( for all <0.0001).
AI-CAC automated LA volumetry and CMRI LA volume measured by human experts similarly predicted incident AF and stroke over 15 years. Further studies to investigate the clinical utility of AI-CAC for AF and stroke prediction are warranted.
与阿加斯顿冠状动脉钙化（CAC）评分相比，人工智能冠状动脉钙化（AI-CAC）能提供更具可操作性的信息。我们最近在动脉粥样硬化多族裔研究（MESA）中表明，AI-CAC自动左心房（LA）容积测量能够早在1年前就预测心房颤动（AF）。
在本研究中，作者评估了AI-CAC LA容积测量与人类专家使用心脏磁共振成像（CMRI）测量的LA容积在预测AF和中风方面的性能，并将它们与基因组流行病学心房颤动心脏与衰老研究队列（CHARGE-AF）风险评分、阿加斯顿评分和N末端B型利钠肽原（NT-proBNP）进行比较。
我们使用了来自3552名无症状个体（52.2%为女性，年龄61.7±10.2岁）的15年结局数据，这些个体在MESA基线检查时同时接受了CAC扫描和CMRI检查。CMRI LA容积先前由人类专家测量。NT-proBNP、CHARGE-AF风险评分和阿加斯顿评分的数据来自MESA。使用时间依赖性曲线下面积评估辨别能力。
在15年的随访中，发生了562例AF和140例中风。AI-CAC与CMRI容积测量预测AF（0.802对0.798）和中风（0.762对0.751）的曲线下面积无显著差异。当添加到CHARGE-AF风险评分（0.23）、NT-proBNP（0.37，0.37）和阿加斯顿评分（0.44）时，AI-CAC LA显著改善了预测5年AF的连续净重新分类指数（所有P<0.0001）。
AI-CAC自动LA容积测量和人类专家测量的CMRI LA容积在15年中对AF和中风的预测相似。有必要进行进一步研究以探讨AI-CAC在AF和中风预测中的临床应用价值。
