Setting up Octavius

Phantom details

The phantom consists of four main parts: The upper half with the eyes is used in both assemblies. Then comes a solid slab, a cut-out slab which holds the ARRAY (see first photo), and a lower part with the mouth drawing on the front side.

Only the lower part with the mouth drawing is changed when switching from the CT assembly to the LINAC assembly.

Instead of the ARRAY, up to nine ionization chambers (PTW 0.125 cm³ thimble chambers, e.g. 31002) can be inserted simultaneously into the phantom.

The LINAC part is used for measurements with the ARRAY. It contains the air cavity which compensates the reduced sensitivity of the ARRAY if irradiated from the back (see also Part I).

The lower part marked with "CT" is mainly needed for the CT scan and is otherwise kept in the storage room.

Note that the CT assembly is also chosen when measuring with chambers instead of the ARRAY. Conventional thimble chambers like the 31002 don't need sensitivity compensation from the back side (which would be hard to find for a cylindrical chamber anyway)!

Parameter	Data
Material	Polystyrol
Density	1.04 g/cm3
Height	320 mm
Width	320 mm
Length	320 mm
Weight (LINAC)	approx. 24 kg
Weight (CT)	approx. 26 kg

High-quality velcro strips hold the phantom together when it is rotated. We do not use the strips in the horizontal setup shown on the first few photos (0° rotation), because nothing can "happen" here. But during rotated setups, especially at 90°, the phantom will fall apart without the velcro strips!

(In case you ask yourself what "velcro strips" are, this should give you an idea. It's a physicist's explanation!)

Couch roll

When setting up the phantom on the linac couch, don't forget to check that the couchtop is also level in the lateral direction. Especially if there are carbon fiber inserts that can easily be removed, there may be significant lateral inclination (couch roll), which should be compensated.

(click to enlarge)

Trouble with CT Scanners

Recently we scanned the phantom in the 90° orientation:

(click to show full plan)

In the Eclipse reconstruction, we measured right angles which should be 90°, of course! We were surprised that they were actually something like 89.4° or 90.7°. The square became a parallelogram.

Maybe there was something wrong with the scanner's gantry tilt. The DICOM slices are sent as transversal (otherwise Eclipse would not accept them), but were they really transversal?

If you have any idea what the reason could be, share your thoughts with us!

Such things are never seen during normal patient scans or even with the ARRAY at 0° rotation. But with the ARRAY at 90°, the regular grid defined by the ARRAY's chambers makes it easy to check this aspect of scanner geometry.

Looks like the ARRAY, due to its grid-like structure, also serves as a nice scanner QA tool.

If you have the ARRAY, you might want to try this out and check your scanner!

We first put everything on the couch and take care that the couch is centered laterally. We do not want to shift the couch in the lateral direction to align Octavius, because then the couch rails would also be shifted. This is not an issue during perpendicular field-by-field verification, but here it is. The couch needs to be centered, and then Octavius needs to be centered on the couch!

The ARRAY is put in the cut-out slab, and some basic alignment is done. The ARRAY interface should be switched on as soon as possible, so the system can warm up.

(click to enlarge)

After the parts are put together, Octavius can be aligned with the lasers. If the couch is level (or rather: if the longitudinal inclination is the same on the linac couch and on the CT couch!), Octavius can be used without "feet":

(click to enlarge)

The small air gap which is visible in the next photo right above the ARRAY will be explained elsewhere.

(click to enlarge)

Octavius has feet which makes it possible to align the phantom very precisely (the feet can be used in all eight possible orientations):

(click to enlarge)

Here is a setup with the ARRAY rotated by 90°. In this case, the feet should be used, because otherwise Octavius would rest on the velcro strips (maybe it's ok, but we never tried it out):

(click to enlarge)

Needless to say: if this setup is chosen, the CT scan also has to be performed with the feet included. During measurements they will not be in the beam, but if the geometry is not consistent, the distance between phantom and couchtop will be different.

Most of the time, the unrotated orientation is chosen, where setup is straightforward, quick and easy. Handling the 24 kg phantom when you are alone in the late afternoon (the physicists typical situation) requires some skills, especially when the phantom's feet should also be included in the setup.

This is a phantom for real men (and strong women), not for weaklings!

The dosimetric performance is discussed here.

back to ARRAY main page