International journal of cardiology · 2026
Atrial myopathy subtypes: Association with and prediction of incident atrial fibrillation-The atherosclerosis risk in communities study (ARIC).
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Moser Ethan D, He Ruoyu, Wang Wendy, Van't Hof Jeremy R, Sun Daokun, Zhang Chunxiao, Zhang Michael J, Chan Lap Sum, Solomon Scott D, Shah Amil M, Pan Wei, Chen Lin Yee
Аннотация
BACKGROUND: Atrial myopathy is a heterogeneous condition associated with atrial fibrillation (AF). This study aimed to identify atrial myopathy subtypes using unsupervised machine learning and assess their association with incident AF.
METHODS: We analyzed 1414 ARIC participants with atrial myopathy, defined by left atrial reservoir strain, and no prevalent AF. Using LASSO for selection among 52 clinical, demographic, ECG, and echocardiographic candidate variables we identified clusters via mClust. A multivariate Cox regression model was used to evaluate the association between clusters and incident AF.
RESULTS: In total, 1414 participants with atrial myopathy were included in the clustering analysis. Three clusters were identified in the clustering analysis based on 26 selected variables. Over a mean follow-up of 6.3 years, 712 incident AF cases occurred. Compared to participants without atrial myopathy (n = 3790), Clusters 1, 2, and 3 showed significantly higher AF risk, with hazard ratios of 2.26, 3.22, and 4.71, respectively. Integrating these clusters into the CHARGE-AF model improved the C-statistic from 0.68 to 0.72.
CONCLUSION: Clusters identified three unique atrial myopathy phenotypes with differential AF risks. These clusters elucidate the heterogeneity of atrial myopathy and enhance AF risk prediction beyond traditional clinical variables.
METHODS: We analyzed 1414 ARIC participants with atrial myopathy, defined by left atrial reservoir strain, and no prevalent AF. Using LASSO for selection among 52 clinical, demographic, ECG, and echocardiographic candidate variables we identified clusters via mClust. A multivariate Cox regression model was used to evaluate the association between clusters and incident AF.
RESULTS: In total, 1414 participants with atrial myopathy were included in the clustering analysis. Three clusters were identified in the clustering analysis based on 26 selected variables. Over a mean follow-up of 6.3 years, 712 incident AF cases occurred. Compared to participants without atrial myopathy (n = 3790), Clusters 1, 2, and 3 showed significantly higher AF risk, with hazard ratios of 2.26, 3.22, and 4.71, respectively. Integrating these clusters into the CHARGE-AF model improved the C-statistic from 0.68 to 0.72.
CONCLUSION: Clusters identified three unique atrial myopathy phenotypes with differential AF risks. These clusters elucidate the heterogeneity of atrial myopathy and enhance AF risk prediction beyond traditional clinical variables.