If you don't trust your MLC, there are not many possibilities that you have for making sure that during IMRT the fluence patterns are delivered correctly on a daily basis. Of course, one could argue that you shouldn't do IMRT if you don't trust your machine. But depending on your linac/MLC model(1) and your degree of confidence, it may be desirable to have permanent in-vivo monitoring of fluence/dose delivery.

One way to accomplish this on Varian machines is to analyse the Dynamic MLC logfiles (dynalog files) after the treatment. The principle was discussed years ago with a recent update. But this cannot be done on the linacs of all three vendors. The advantage of this approach is that it doesn't interfere with the treatment at all: nobody has to enter the room to attach some equipment to the linac head before the treatment or to remove it afterwards.

The second best approach in terms of interfering with the treatment is probably the wireless DAVID system from PTW Freiburg (see sidebar). The measurement wires, one for each leaf pair, run along the mid-line of the leaves. The wires are placed between two windows which are at a potential of +400 V. During treatment of a MLC field, variable lengths of the wires get exposed to radiation. The ion pairs are collected by the wires. The principle is similar to a thimble ionization chamber with a long central electrode. The signal of each wire can be interpreted as dose length product. Due to the fact that the sensitive volume of each wire is open on the sides, there is some "cross-talk". But in the context of constancy checks, this is no problem: one refers to a reference measurement.

Due to its transparent chamber design, DAVID may stay attached to the collimator the whole day (unless the slot is needed for other things like hard wedges). PTW asserts that the visibility of the light field is not restricted, and that the dosimetric absorption of the transmission chamber can be compensated with a tray factor.

DAVID on carrying case
(The device - here in it's carrying case - is slim and lightweight.)

DAVID on Clinac
(DAVID is inserted into the slot where the hard wedges and the front pointer go.)

KFJ testing

For several weeks, we tested the device on our Varian linacs (Clinac 2300C/D with 80-leaf MLC). On one hand, we looked at DAVID as a "passive device". We made dosimetric measurements regarding attenuation, increase of surface dose, etc.

On the other hand we used DAVID actively. We made a usability testing and also looked at potential other usages than the ones intended by PTW.

DAVID (passive)

Putting a device like DAVID into the beam modifies certain beam quality parameters.

DAVID on Clinac
(The good visibility of the light field is demonstrated on the PortalVision detector.)

DAVID (active)


(1)... thinking of the only linac vendor who also produces coffeemakers.