首页> 外国专利> METHOD AND SYSTEM FOR PURELY GEOMETRIC MACHINE LEARNING BASED FRACTIONAL FLOW RESERVE

METHOD AND SYSTEM FOR PURELY GEOMETRIC MACHINE LEARNING BASED FRACTIONAL FLOW RESERVE

机译:基于纯几何机器学习的分数流储备的方法和系统

摘要

A kind of method and system such as blood flow reserve (FFR), is disclosed for measuring hemodynamics index in the coronary artery of the patient of interested position. The medical image of patient is received. Patient-specific coronary arterial tree geometry is by the medical data from patient. Geometrical characteristic, the patient of geometrical characteristic are extracted from patient-specific coronary arterial tree. A kind of hemodynamic index, such as FFR, position of interest is based on patient-specific coronary arterial tree and extracts geometrical characteristic in calculating, is had great significance using trained machine learning based on agent model. The machine learning have great significance based on agent model be trained to based on Extraction of Geometrical Features from the geometry for being synthetically produced coronary arterial tree.
机译:公开了一种诸如血流储备(FFR)的方法和系统,用于测量感兴趣位置的患者的冠状动脉中的血液动力学指数。接收患者的医学图像。患者特定的冠状动脉树的几何形状来自患者的医学数据。几何特征,患者的几何特征是从患者特定的冠状动脉树中提取的。一种基于患者特定冠状动脉树的血液动力学指标(如FFR)基于感兴趣的冠状动脉树,并提取几何特征进行计算,对于基于Agent模型的受过训练的机器学习具有重要意义。机器学习对基于Agent模型的训练对基于几何特征的几何特征提取对于被人工合成的冠状动脉树的训练具有重要意义。

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号