European heart journal. Cardiovascular Imaging · 2025
Development and validation of a radiomics model for detecting cardiac amyloidosis at coronary computed tomography angiography.
СтатьяРандомизированное исследованиеВалидационное исследованиеНазвание 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.
Meng Qingchao, Zhao Li, Sun Xiaoxin, Wang Yang, Yu Lu, Schoepf U Joseph, Varga-Szemes Akos, Kravchenko Dmitrij, Wang Yining, Liu Hui, Zhang Yan, Xu Lei, Yu Xianbo, Guo Yubo, Chen Jiayu, Feng Dong, Bo Kairui, Gao Yang, Lu Bin
Аннотация
AIMS: To investigate the diagnostic performance of computed tomography (CT)-based radiomics in detecting cardiac amyloidosis (CA) in patients with diffuse myocardial thickening.
METHODS AND RESULTS: Patients with diffuse myocardial thickening who underwent coronary CT angiography were retrospectively enrolled from five hospitals. Patients from one hospital were randomly divided into training and internal test cohorts at a 7:3 ratio, and the other four hospitals constituted the external test cohort. The diagnosis of CA followed established guidelines. Regions of interest of myocardium were delineated to extract radiomics features to construct the radiomics model, and myocardial CT attenuation was measured. The diagnostic performance and clinical utility of the radiomics model and myocardial CT attenuation were compared with the area under the curve and decision curve analysis. The correlation between radiomics score and left ventricular function was analysed. A total of 378 patients (median age, 57 years; 257 men) were enrolled. Ten features were selected to construct the radiomics model. The areas under the curve of radiomics model were significantly higher than myocardial CT attenuation in the training (0.95 vs. 0.58, P < 0.001), internal test (0.95 vs. 0.59, P < 0.001), and external test cohorts (0.91 vs. 0.64, P < 0.001). Decision curve analysis indicated the radiomics model provided a greater net benefit than myocardial CT attenuation across cohorts. Radiomics scores were correlated with N-terminal proB-type natriuretic peptide and left ventricular diastolic diameter across cohorts (P < 0.05).
CONCLUSION: The radiomics model exhibited good diagnostic performance for CA detection in patients with hypertrophic phenotypes, outperforming myocardial CT attenuation.
METHODS AND RESULTS: Patients with diffuse myocardial thickening who underwent coronary CT angiography were retrospectively enrolled from five hospitals. Patients from one hospital were randomly divided into training and internal test cohorts at a 7:3 ratio, and the other four hospitals constituted the external test cohort. The diagnosis of CA followed established guidelines. Regions of interest of myocardium were delineated to extract radiomics features to construct the radiomics model, and myocardial CT attenuation was measured. The diagnostic performance and clinical utility of the radiomics model and myocardial CT attenuation were compared with the area under the curve and decision curve analysis. The correlation between radiomics score and left ventricular function was analysed. A total of 378 patients (median age, 57 years; 257 men) were enrolled. Ten features were selected to construct the radiomics model. The areas under the curve of radiomics model were significantly higher than myocardial CT attenuation in the training (0.95 vs. 0.58, P < 0.001), internal test (0.95 vs. 0.59, P < 0.001), and external test cohorts (0.91 vs. 0.64, P < 0.001). Decision curve analysis indicated the radiomics model provided a greater net benefit than myocardial CT attenuation across cohorts. Radiomics scores were correlated with N-terminal proB-type natriuretic peptide and left ventricular diastolic diameter across cohorts (P < 0.05).
CONCLUSION: The radiomics model exhibited good diagnostic performance for CA detection in patients with hypertrophic phenotypes, outperforming myocardial CT attenuation.