论文部分内容阅读
摘要目的探讨能否通过机器学习三维心脏收缩运动模式预测肺动脉高压所致右心室心功能衰竭病人生存期及发生机制。材料与方法本研究经研究伦理委员会批准,获得所有参与者书面知情同意。256例新近诊断为肺动脉高压的病人[女143例,平均年龄(63±17)岁]行心脏MRI检查,右心导管检查及6 min步行试验,平均随访4年。利用短轴电影图像半自动分割法创建一个右心室运动的三维模型,应用监督主要成分的方法来识别最能预测生存期的收缩期运动模
Abstract Objective To investigate whether the survival time and its mechanism of patients with right ventricular failure caused by pulmonary hypertension can be predicted by machine learning three-dimensional systolic motion model. Materials and Methods The study was approved by Research Ethics Committee with written informed consent from all participants. A total of 256 newly diagnosed pulmonary hypertension patients (143 women, mean age 63 ± 17 years) underwent cardiac MRI, right heart catheterization and 6-min walking test, with a mean follow-up of 4 years. Semi-automatic segmentation using short-axis movie images to create a three-dimensional model of right ventricular motion, the application of the main components of surveillance methods to identify the most predictive of survival systolic motion mode