Example of an intervention using a catheter to release rectum gas. The first row shows the transverse and sagittal cross sections on the planning CT. The second and third rows show the same cross sections before and after intervention, respectively. The planning CT contours for the prostate-bed, bladder, and rectum are outlined for comparison. The catheter is visible as a diagonal line on the planning CT and post-intervention sagittal cross section.
Patient mean shift in all three directions for pre-intervention data (first row), post-intervention data (second row), and their difference (third row). The lines indicate the overall population mean shift (solid horizontal line) and standard deviation (dashed horizontal lines).
Scatterplots of the position correlations between the pre- and post-intervention data in all directions. The coordinates for each point correspond to its pre-intervention shift and its post-intervention shift. The axes of the ellipses are the standard deviations of the pre-intervention data and post-intervention data for each direction. Points inside the white and gray areas represent posterior and anterior shifts, respectively, of the target volume with respect to the pre-intervention position.
Histogram of the mean 3D shift values. (a) Pre-intervention data and (b) post-intervention data.
Effect of rectum gas on the target area deformations. The first row shows transverse, coronal, and sagittal cross sections from the planning CT. The second and third rows show similar data from the pre- and post-intervention scans, respectively. The planning CT prostate-bed, bladder, and the rectum volumes are contoured. For this fraction, the intervention consisted in having the patient go to the restroom to deflate the rectum. The second scan was taken half an hour after the first scan.
Mean shift and standard deviation in each direction for pre- and post-intervention data. The mean 3D vector value and its standard deviation are also shown.
Relative shifts between the post- and pre-intervention positions for all kinds of interventions as well as excess rectal content and excess bladder interventions separately.
-values of the F-test for equality of variance and of the Wilcoxon test for equality of distribution between the different pairs of data sets (pre- and post-intervention).
Percentage of pre- and post-intervention shifts greater than 5 mm in the LR, SI, and AP directions for the pre- and post-intervention localization data.
Systematic error , random error , and uniform CTV-to-PTV margin suggested by the pre- and post-intervention data sets. Two pre-intervention results are reported corresponding to different localization methods that either take into account CTV and OARs simultaneously (CTV/OAR) or only consider CTV overlap (CTV only).
Article metrics loading...
Full text loading...