SU-D-18C-07: Framework for Time-Dependency Analysis of MRI Data to Visualize and Quantify Lung Ventilation by Using Oxygen as Contrast Media




To visualize and quantify lung ventilation by analyzing MRI data time-dependent an eligible framework is capable to use the contrast enhancing characteristic of oxygen in functional lung MRI.


A framework was developed and programmed by using the tool Fiji, which is an image processing package based on ImageJ. The analysis was performed on previously recorded 1.5T magnetic resonance images of narcotized and mechanically ventilated pigs in both dorsal and transversal plane, with a slice thickness of 8 mm and 32 pictures per slice. The oxygen rate of the pigs' breathing was alternated periodically from 21 to 100 percent and backwards within an eight seconds interval. Several lung areas were blocked mechanically. After importing the MR sequence into Fiji the framework could be started automatically with pre-set or manually adjusted parameters. The framework analyzed all pixels of each picture temporally to detect the variation of gray values related to the temporal change in oxygen rate.


The software is able to detect time-dependent changes in signal intensity during dynamic examination of the lung. Change of pixel intensities and time-to-peak of the output signal showed a significant relation between oxygen rate and gray value of the stored data and the generated images and curves. The blocked lung areas could be identified in the resultimages since pixels located in these regions representing less signal in contrast with other regions.


This work confirmed the ability of oxygen to be used as contrast media in magnetic resonance imaging. The developed and tested framework for time-dependent analysis of MRI data has demonstrated the significance of a quantification of lung ventilation based on oxygen-induced change in output signal.