All patients underwent CT and MRI imaging studies under a prospective protocol approved by the institutional review board (IRB). The time interval between CT and MRI studies was less than 1 week. Patients were in a head-first supine position with arms up. CT scans were performed on a 16-slice scanner (Optima CT580™; GE Healthcare, Milwaukee, WI) equipped with the Real-time Position Management (RPM™) system (Varian Medical Systems, Palo Alto, CA) and AdvantageSim™ 4D software (GE Healthcare, Milwaukee, WI). 4D-CT scans were performed in cine mode with the following parameters: 120 kV, variable mA, gantry rotation period of 1 second, and slice thickness of 2.5mm. The cine duration time was set to be the patient’s breathing period plus 1.5 seconds.
MRI scans were performed on a 3.0 Tesla system (MAGNETOM Verio™, Siemens Healthcare, Germany) equipped with a 6-channel body matrix coil. No special immobilization was used. The imaging protocol included an 8-minute free-breathing SG-KS-4D-MRI scan (prescribed isotropic spatial resolution = 1.56 mm, a cubic imaging volume of 300×300×300 mm3, flip angle=10°, TR/echo time (TE)=5.8/2.6ms, readout band width=399Hz/pixel) and two 1.2 minute free-breathing cine 2D-MRI spoiled gradient recalled echo (GRE) scans with orthogonal imaging planes traversing the tumor (351 ms/frame, 1.56-mm in-plane spatial resolution, 8-mm slice thickness).
MRI scans were performed on a 3.0 Tesla system (MAGNETOM Verio™, Siemens Healthcare, Germany) equipped with a 6-channel body matrix coil. No special immobilization was used. The imaging protocol included an 8-minute free-breathing SG-KS-4D-MRI scan (prescribed isotropic spatial resolution = 1.56 mm, a cubic imaging volume of 300×300×300 mm3, flip angle=10°, TR/echo time (TE)=5.8/2.6ms, readout band width=399Hz/pixel) and two 1.2 minute free-breathing cine 2D-MRI spoiled gradient recalled echo (GRE) scans with orthogonal imaging planes traversing the tumor (351 ms/frame, 1.56-mm in-plane spatial resolution, 8-mm slice thickness).