Portal Dosimetry - Dosimetric Problems ...

...or rather "Challenges!"

The first thing I noticed when I evaluated some DICOM images acquired with the aS500 was that the beam profiles seemed to "hang" towards the Gantry (Fig.1).

Fig.1: Hanging inplane profiles for a 10x10 field. Gantry is on the right side.

Everybody was alarmed because we first thought the beam was really that asymmetric! But water phantom measurements quickly revealed that the true beam symmetry was nearly perfect. This is valid for all screenshots shown on this page: the real beam was always good ...

So the effect was probably originating from the image acquisition system. This was also confirmed when the images of the two identical linacs were compared: they showed the same effect.
The asymmetry is a little stronger for 6MV than for 15MV. In Crossplane direction, symmetry is ok. The effect depends on field size. For 25x25, or 27x36 (the calibration field size), the inplane symmetry is much better, but for smaller field sizes ... see Fig.2 and 3, comparing three 6MV inplane profiles. Gantry is on the right for all inplane profiles shown.

Fig.2: 6 MV inplane profiles for three different field sizes. For the 10x10 field, the profile hangs.

Fig.3: Beam Analysis of the three profiles shown in Fig.2. For 10x10, the measured beam asymmetry is nearly 3%.
If dosimetry calibration is done without the beam profile correction, profiles are more or less flat on the plateau. The asymmetry is still there. The effect is strongest for long fields of narrow width. Fig.4 shows a 6MV field, field size is 27x3 cm..

Fig.4: A narrow and long field (see insert on the left). The imager was calibrated without beam profile correction. The plateau therefore should be perfectly flat.

Fig.5 shows the plateau region of Fig.4 zoomed. The high energy image looks similar (not shown).

Fig.5: Zoomed plateau in the 6MV image.

One more image shall be shown, demonstrating that the effect does not depend on the number of MU applied. In Fig.6, three measured inplane profiles, acquired at 3, 30 and 300 MU are shown. However, the profile measured at 3 MU has more noise.

Fig.6: Inplane profiles of a 6 MV, 27x36 cm field taken at 3 (green), 30 (blue) and 300 MU (red).

The "hanging profiles" problem is one of the things Varian needs to solve. The situation is identical for IDU11 and IDU20.

Nevertheless, the Varian Portal Dosimetry system is very useful and is routinely used for IMRT verification in our department. To reduce the inplane effect, it often helps to rotate the collimator to 90°, if the fields are longer than wide.

The following table sums up some general advantages and disadvantages of the Portal Dosimetry system:

Pros Cons
Very fast. No Phantom setup required
Not measuring dose, rather fluence
Controlled by ARIA. No setup error possible
TPS uses different algorithms for 3D-dose and PDIP
Documented by ARIA. No lost data
No cumulation of subfields during large-field IMRT possible
High resolution digital image (512x384 or 1024x768)
No plan verification, only field verification
40x30cm measurement area
...

back to Overview