Pulmonary Perfused Blood Volume with Dual-Energy CT as Surrogate for Pulmonary Perfusion Assessed with Dynamic Multidetector CT

© RSNA, 2012

Supplemental Figures

Figure E1. Example CT images and time-intensity curves from lung inflation model (top) and balloon occlusion model (bottom). Time (in seconds) is plotted along the x-axis, and Hounsfield units are plotted along the y-axis. A gravitational gradient in the CT-derived PBV is clearly evident in the lung inflation model, where regions A–E ROIs represent parenchymal regions in dependent to nondependent lung regions. Placement of the occlusion balloon has severely reduced the CT-derived PBF in the left lung (bottom, curves A and E). ROIs B, C, and D show a gravitationally dependent reduction in the area under the curve, whereby B is dependent and D is nondependent. In addition to the time-intensity curves shown here, an arterial input curve (not shown) from an ROI placed in the region of a more central pulmonary artery was used to calculate CT-derived PBF, as described in Materials and Methods.

Figure E2. Screenshot of the dual-energy application used to calculate PBV. Images on left are sagittal and transverse views of lung in gray scale with a color overlay of CT-derived PBV values. Upper right panel shows the parameter box used to input the energy-dependent three-material decomposition attenuation properties used to calculate PBV. A three-dimensional visualization of the pulmonary vasculature is visible in the bottom right panel.