Electrical impedance tomography (EIT) is a promising non-invasive, radiation-free imaging modality. Using EIT-derived index Center of ventilation (CoV), ventral-to-dorsal shifts in distribution of lung ventilation can be assessed. The methods of CoV calculation differ among authors and so does the segmentation of EIT images from which the CoV is calculated. The aim of this study is to compare the values of CoV obtained using different algorithms, applied in variously segmented EIT images. An animal trial (n=4) with anesthetized mechanically ventilated pigs was conducted. In one animal, acute respiratory distress syndrome (ARDS) was induced by repeated whole lung lavage. Incremental steps in positive end-expiratory pressure (PEEP), each with a value of 5 cmH2O (or 4 cmH2O in the ARDS model), were performed to reach total PEEP level of 25 cmH2O (or 22 cmH2O in the ARDS model). EIT data were acquired continuously during this PEEP trial. From each PEEP level, 30 tidal variation (TV) images were used for analysis. Functional regions of interest (ROI) were defined based on the standard deviation (SD) of pixel values, using threshold 15%–35% of maximum pixel SD. The results of this study show that there might be statistically significant differences between the values obtained using different methods for calculation of CoV. The differences occured in healthy animals as well as in the ARDS model. Both investigated algorithms are relatively insensitive to the image segmentation.
Center of Ventilation Center of Gravity Electrical Impedance Tomography EIT Region of interest