Phantom designed for measurements of contrast-to-noise ratio. Inserts mimicking bodily tissues were implanted in a head-size water cylinder. The image on the left is a transverse slice of the MVCBCT reconstruction and shows the regions-of-interest that were defined for CNR measurements.
Stability of two MVCBCT systems (linacs 1 and 2) geometry in rotation. The plots show mean projection position of the treatment isocenter on the flat panel using projection matrices acquired over a period of for all gantry angles of the acquisition. The curves plotted with the mean position in rotation represents two standard deviations for the period.
CNR performance as a function of MVCBCT exposure (MU) for three different imaging protocols identified: baseline, and and diffusion. The CNR performances are reported for three electron densities relative to water [ (dense bone), 1.17 (trabecular bone), and 1.043 (muscle)]. Error bars represent the standard deviation of the CNR for axial images analyzed over a length of .
Axial images of the phantom [Fig. 2] showing the CNR performance as a function of MVCBCT exposure (MU) for two different reconstruction protocols: baseline and .
Techincal characteristics of MVCBCT imaging. The bold font specifies the typical values used in the baseline image acquisition and reconstruction protocol.
Flat panel detector intensity statistics for flood field MVCBCT projection images acquired at different exposure for two treatment units (linacs 1 and 2). The mean unattenuated signal per projection for a given exposure was obtained from a large region of interest centered on the projection images. The standard deviation and the minimum and the maximum signal values are presented as a ratio of the mean signal. The error bars correspond to the standard deviation of all air MVCBCT projection measurements acquired over a period of .
Uniformity of the water cylinder reconstructed with MVCBCT. The uniformity is reported for the axial and sagittal central planes as a percentage of the mean signal for the whole water cylinder. Uniformity is reported for different combinations of craniocaudal imaging lengths and gains performed either in air or using a lead plate. The uniformity is also reported for two postprocessing correction methods.
Spatial resolution performance for different reconstruction protocols. The full width at half maximum of the point-spread function is reported. Only the specified protocol component is charged compared to the baseline specified in Table I.
CNR performance for different reconstruction protocols. Only the specified protocol component is changed compared to the baseline in Table I. The CNR performances are reported for three electron density differences relative to water (rED).
Article metrics loading...
Full text loading...