Light speed volume ct

Computed tomography examinations were performed using the following parameters: 120 kV; 240–260 mAs; collimation, 64 × 0.6; and slice thickness, 5 mm (Brilliance iCT, Philips Healthcare, Amsterdam, the Netherlands; Light Speed Volume CT, GE Healthcare, Chicago, Il). Computed tomography enhancement was performed after a 70-second delay after intravenous administration of 1.5 mL/kg of iodinated contrast (100 mL of 370 mg J/mL iopromide; Bayer Schering Pharma, Berlin, Germany) using a power injector (Ulrich CT Plus 150; Ulrich Medical, Chesterfield, Mo) at a rate of 2.5 mL/s through the antecubital vein.
Two experienced radiologists (Y.L. and P.Y., with 6 and 10 years of CT experience, respectively) cooperatively delineated the regions of interest (ROIs) on both the CT and CTE images along the boundary of each tumor in each axial slice using the uAI Research Portal^{22 (link)} (Shanghai United Imaging Intelligence Co, Ltd, Shanghai, China). The ROI decisions were achieved through consensus. The ROIs were automatically expanded by 3 mm as the volume of interest for peritumoral evaluation. Four ROIs (Pre-CT, Pre-CT_{+3 mm}, CTE, and CTE_{+3 mm}) were defined in axial, coronal, and sagittal CT images, as shown in Figure 1.

A preinterventional CT was performed per clinical indication on GE Lightspeed Volume CT, slice thickness 0.625 mm, prospectively acquired in mid-diastole. The relevant ethics committee waived informed consent for further use of the images. This was a proof of concept study using a patient-based rather than an artificial virtual environment, without interventions in humans. Only one set of patient data was used to create the virtual environment for stent testing at this time, analogous to conventional bench testing of new stent designs in one artificial virtual environment. Patient data were chosen based on the following criteria: adult, aortic coarctation in the typical location just distal to the left subclavian artery origin and good quality CT scan. DICOM data from the aortic and stent CTs were imported into the virtual environment. The distal end of the aortic model was chosen to be approximately 5 cm below the prospective distal end of the stent to ensure that boundary conditions did not influence implantation.