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.
Localization of deformable tumors from short-arc projections using Bayesian estimation
Rent this article for
View: Figures


Image of FIG. 1.
FIG. 1.

Schematic illustration of the stochastic triangulation model with the patient contour represented by an ellipse. (a) The initial setup error after laser alignment of the skin marks, represented by the displacements of the reference point (random variable ). (b) The displacement of each marker due to the deformations inside the patient's body (random variable ) relative to the current reference point position . (c) The geometric projection at imaging angle θ n of the marker positions with uncertainties in the radiographs (random variable ).

Image of FIG. 2.
FIG. 2.

(Top) The CIRS IMRT thorax phantom with three 1.5 mm marker balls which can be manually displaced to each other. (Bottom) An example cone beam projection, acquired with the Elekta Synergy® XVI system, with additional marker extraction representing the input image q n for the estimation.

Image of FIG. 3.
FIG. 3.

Differences of image qualities of the prostate case with four markers. (Left) The anterior-posterior view with high image quality. (Right) The lateral view with low image quality.

Image of FIG. 4.
FIG. 4.

The scheme of 18 arc segments with 5° arc lengths and 10 projections each. The numbering of the arc segments is chosen to correspond to the results.

Image of FIG. 5.
FIG. 5.

The residual errors of 20 phantom setups (as described in Sec. II D) with the initial setup and after correction based on CM and ML in a 40° and 10° imaging arc length using 20 projections each. (Top) Errors for every setup in order of the highest (about 3.1 cm) to the lowest initial setup error (about 0.6 cm) which is presented as black triangles. All other residual errors are presented by symbols in the legend. (Middle) The corresponding histogram of residual errors with the initial setup error represented as white bars and the other results in gray to black shaded bars. (Bottom) The histograms for x-, y- and z-direction with the same shading as in the histogram of the middle row.

Image of FIG. 6.
FIG. 6.

The residual errors of the 18 imaging directions for the BE approach, as described in Fig. 4. (Top) The residual error of each arc segment. (Bottom) The histogram of the residual errors of the arc segments all around the patient subdivided into lateral and nonlateral views.


Article metrics loading...


Full text loading...

This is a required field
Please enter a valid email address
752b84549af89a08dbdd7fdb8b9568b5 journal.articlezxybnytfddd
Scitation: Localization of deformable tumors from short-arc projections using Bayesian estimation