Department of Medicine, Beth Israel Deaconess Medical Centre and Harvard Medical School, 330 Brookline Avenue, Boston, MA 02215, USA.
Department of Computer Science, Technical University of Munich, Arcisstraße 21, 80333 Munich, Germany.
Eur Heart J Cardiovasc Imaging. 2022 Mar 22;23(4):532-542. doi: 10.1093/ehjci/jeab056.
Cardiovascular magnetic resonance (CMR) with late-gadolinium enhancement (LGE) is increasingly being used in hypertrophic cardiomyopathy (HCM) for diagnosis, risk stratification, and monitoring. However, recent data demonstrating brain gadolinium deposits have raised safety concerns. We developed and validated a machine-learning (ML) method that incorporates features extracted from cine to identify HCM patients without fibrosis in whom gadolinium can be avoided.
An XGBoost ML model was developed using regional wall thickness and thickening, and radiomic features of myocardial signal intensity, texture, size, and shape from cine. A CMR dataset containing 1099 HCM patients collected using 1.5T CMR scanners from different vendors and centres was used for model development (n=882) and validation (n=217). Among the 2613 radiomic features, we identified 7 features that provided best discrimination between +LGE and -LGE using 10-fold stratified cross-validation in the development cohort. Subsequently, an XGBoost model was developed using these radiomic features, regional wall thickness and thickening. In the independent validation cohort, the ML model yielded an area under the curve of 0.83 (95% CI: 0.77-0.89), sensitivity of 91%, specificity of 62%, F1-score of 77%, true negatives rate (TNR) of 34%, and negative predictive value (NPV) of 89%. Optimization for sensitivity provided sensitivity of 96%, F2-score of 83%, TNR of 19% and NPV of 91%; false negatives halved from 4% to 2%.
An ML model incorporating novel radiomic markers of myocardium from cine can rule-out myocardial fibrosis in one-third of HCM patients referred for CMR reducing unnecessary gadolinium administration.
心血管磁共振（CMR）与晚期钆增强（LGE）越来越多地用于肥厚型心肌病（HCM）的诊断、风险分层和监测。然而，最近的数据表明，脑部钆沉积引起了安全担忧。我们开发并验证了一种机器学习（ML）方法，该方法结合了从电影中提取的特征，以识别没有纤维化的 HCM 患者，从而避免使用钆。
使用 XGBoost ML 模型，从电影中提取区域性壁厚度和增厚、心肌信号强度、纹理、大小和形状的放射组学特征。使用来自不同供应商和中心的 1.5T CMR 扫描仪收集了包含 1099 例 HCM 患者的 CMR 数据集，用于模型开发（n=882）和验证（n=217）。在 2613 个放射组学特征中，我们在开发队列中使用 10 折分层交叉验证，确定了 7 个特征，这些特征在+LGE 和-LGE 之间提供了最佳区分。随后，使用这些放射组学特征、区域性壁厚度和增厚开发了 XGBoost 模型。在独立验证队列中，ML 模型的曲线下面积为 0.83（95%CI：0.77-0.89），灵敏度为 91%，特异性为 62%，F1 评分为 77%，真阴性率（TNR）为 34%，阴性预测值（NPV）为 89%。为了提高灵敏度进行了优化，灵敏度为 96%，F2 评分为 83%，TNR 为 19%，NPV 为 91%；假阴性率从 4%降至 2%。
一种结合电影中新型心肌放射组学标志物的 ML 模型可以排除三分之一接受 CMR 检查的 HCM 患者的心肌纤维化，从而减少不必要的钆剂使用。
