1887
banner image
No data available.
Please log in to see this content.
You have no subscription access to this content.
No metrics data to plot.
The attempt to load metrics for this article has failed.
The attempt to plot a graph for these metrics has failed.
Multicriteria optimization of the spatial dose distribution
Rent:
Rent this article for
USD
10.1118/1.4828840
/content/aapm/journal/medphys/40/12/10.1118/1.4828840
http://aip.metastore.ingenta.com/content/aapm/journal/medphys/40/12/10.1118/1.4828840

Figures

Image of FIG. 1.
FIG. 1.

An illustration of a trade-off represented by isodoses. (a) The scenario includes the PTV (gray), two OAR (white), and the two bold isodose lines representing the lower PTV bound and the upper OAR bound, respectively. (b) It would be preferable to modify the isodose to the dotted shape, which requires relaxation of bounds. (c) A possible result of relaxation with dose–volume constraints. Arrows indicate problematic regions.

Image of FIG. 2.
FIG. 2.

Continuing the example from Fig. 1 to illustrate local relaxation. (a) The actually limiting regions on the isodose are highlighted. (b) Graphical visualization of a reasonable relaxation of the isodose shape. (c) A subsequent optimization could move the isodoses as indicated by the dark gray regions, the desired improvement can be realized only partially, as the constraints limit further progress. The trade-off becomes obvious.

Image of FIG. 3.
FIG. 3.

(a) The subset of voxels of the rectum selected as target (cubes) and the remaining voxels of the rectum (spheres), and a wire- frame rendering of the 33 Gy iso-dose surface. The 36 Gy iso-dose surface covering the PTV voxels (spheres) before (b) and after (c) PTV lower bound relaxation. The arrows indicate where the lower bound for PTV voxels was relaxed, resulting in more voxels not covered by the 36 Gy iso-dose surface. Note that we consider the centroid of the spheres representing the voxels in our constraints.

Image of FIG. 4.
FIG. 4.

Axial (a) and sagittal (b) slices showing the 33 Gy and 36.25 Gy iso-dose lines before (solid) and after (dotted) all relaxation steps. Changes are limited to the selected region on the rectal wall and to the periphery of the bladder.

Image of FIG. 5.
FIG. 5.

DVHs summarizing the dose distributions for the different VOIs and the respective subvolumes for the initial optimization (solid) and after local PTV and bladder relaxation (dashed), respectively. (a) DVHs for PTV, rectum, and bladder. (b) DVHs for the entire rectum and for its target and non-target subvolumes. (c) DVHs for the entire PTV and for its relaxed and non-relaxed subvolumes. (d) DVHs for the entire bladder and for its relaxed and non-relaxed subvolumes. The DVHs illustrate that only the dose bounds in the selected subvolumes are affected by the relaxation steps.

Image of FIG. 6.
FIG. 6.

(a) The desired boost volume (wire-frame) and the final 45.3 Gy iso-dose surface. Axial (b) and sagittal slices (c) showing the 20, 33, 41.5 and 45.3 Gy iso-dose lines before (solid) and after (dashed) relaxation. The dashed black contour in the axial slice illustrates the relaxation of the shell towards the sides.

Image of FIG. 7.
FIG. 7.

DVHs for boost volume, PTV, rectum, bladder for the initial distribution (solid) and the dose distributions after completion of all local relaxation steps (dashed).

Image of FIG. 8.
FIG. 8.

Voxels of SHELL2 represented by spheres where the dose bounds were maintained and by boxes, where the dose bounds were relaxed, and the 4 Gy iso-dose surface: (a) before relaxation SHELL2 contains the iso-dose surface, but some highlighted areas indicate that bounds are limiting the optimization of coverage; (b) after successively relaxing the bounds towards the nasal cavity dose fingers extend in this direction only.

Image of FIG. 9.
FIG. 9.

Axial (a) and sagittal (b) slices showing the 4.0, 7.0, 10.0, and 18.0 Gy isodose lines before (solid line) and after (dashed line) all relaxations.

Image of FIG. 10.
FIG. 10.

DVHs summarizing the dose distributions for the different VOIs and the respective subvolumes for the initial optimization (solid) and after all relaxations (dashed), respectively. (a) DVHs for the entire PTV and for its target and non-target subvolumes. (b) DVHs for the entire optic nerve and for its relaxed and non-relaxed subvolumes. (c) DVHs for the entire brainstem and for its relaxed and non-relaxed subvolumes. (d) DVHs for the entire SHELL2 and for its relaxed and non-relaxed subvolumes. The DVHs illustrate that only the dose bounds in the selected subvolumes are affected by the relaxation steps.

Tables

Generic image for table
TABLE I.

Summary of the actual dose statistics after local trade-off steps for selective sparing of the rectum. The average runtime is denoted by . Note that the initial step represents the baseline for interactive planning.

Generic image for table
TABLE II.

Summary of the actual dose statistics after steps to increase coverage to the boost volume inside the PTV. The average runtime is denoted by . Note that the initial step represents the baseline for interactive planning.

Generic image for table
TABLE III.

Summary of the actual dose statistics after trade-off steps to improve coverage of the PTV. The average runtime is denoted by . Note that the initial step represents the baseline for interactive planning.

Loading

Article metrics loading...

/content/aapm/journal/medphys/40/12/10.1118/1.4828840
2013-11-18
2014-04-16
Loading

Full text loading...

This is a required field
Please enter a valid email address
752b84549af89a08dbdd7fdb8b9568b5 journal.articlezxybnytfddd
Scitation: Multicriteria optimization of the spatial dose distribution
http://aip.metastore.ingenta.com/content/aapm/journal/medphys/40/12/10.1118/1.4828840
10.1118/1.4828840
SEARCH_EXPAND_ITEM