ValidEIT - Validation of regional lung perfusion based on electrical impedance tomography (EIT) by computed tomography (CT) and invasive flow measurement (Swan-Ganz catheter)

点击翻译按钮获取中文摘要

Electrical impedance tomography (EIT) is a non-invasive imaging technique that is successfully used in intensive care medicine for the continuous bed-side monitoring of regional lung ventilation in mechanically ventilated patients. However, since gas exchange only takes place in both adequately ventilated and perfused lung regions, it would be highly desirable to additionally determine regional lung perfusion using EIT. That way, completely new approaches to optimizing ventilation therapy would be conceivable.Regional lung ventilation using EIT is determined based on the ventilation-related signal (VRS) which is caused by respiratory activity. In addition, a pulsatile cardiac-related signal (CRS) can be observed in EIT measurements. However, its exact origin has not yet been conclusively clarified. It is assumed that the CRS in the lung area does - at least partially - result from pulsatile volume changes of the microvascular bed, which would imply a direct connection to regional lung perfusion. Hence, the aim of this research project is to investigate the physiological origin of CRS in order to enable a non-invasive estimation of regional lung perfusion using EIT. In an animal trial, the EIT perfusion estimate from the CRS shall be compared with computed tomography perfusion scans (CT-P) serving as an established reference procedure. Lung perfusion will be modulated regionally and globally by the placement of a balloon catheter at different levels of the pulmonary vascular tree as well as by drug intervention. In addition, a complex finite element model (FEM) of the thorax will be developed in order to perform an isolated investigation of the influences of various factors (conductivity changes, geometric shifts, flow velocities) on the CRS. Model- and signal-based approaches are used to estimate regional lung perfusion. Based on the data of the FEM simulations, deterministic and stochastic CRS models are developed, which are validated on the data of the animal trial. Furthermore, blind and semi-blind source separation as well as deconvolution techniques will be used to analyze the signal origin of the CRS.