Ummmm no we only have one flattening filter per MV energy (at least on a Varian). So enlighten me about the linac with field size dependent flattening filters.

Dear Lindsay and Avinash,

I repat your Ummmm, because I never heard to use different flattening filters for one photon energy. What you find is a field size dependence for the filter factors.

I append a chapter from my dosimetry text book concerning dose distributions and experimental results for flattening filters. It´s written in German but I´m sure you find all the informatios you need. The linacs are Varian.

A small appendix. I remember, that we constructed our own flattening filters for very large fields, because the manufacturers restricted the filed size for the standard filters. There was no dosimetric reason, just matching to the fields.

Hi Lindsay and Hanno,

Thank you for sharing your thoughts and for the document you had shared, its in German but I could translate using GOOGLE translate.

Coming to the discussion,

If we have a 6 MV conventional tube which can deliver max of 300 to 400 MU/min for a 40 cross 40 field with a flattening filter. (This value is with max PRF available). Now to increase the MU/min I would design a different flattening filter for a field of 10 cross 10 which can deliver 800 MU/min (this tube delivers max of 1000 MU/min unflatten) and I could design different flattening filters for different field sizes to get different MU/min.20 X 20 with 600 Mu/min etc.

Please comment on the possibility of this.

I must confess, I don´t understand the problem. I will show you my knowledge. MU are registered before flattening filter, the position of the dose monitor is above the filter for 0 degree gantry. So the existence of the filter does not influence the machine output.

If you determine your MU below the filter (a strange terminolgy), then your filter reduces obviously the dose rate in the water phantom or PMMA phantom, say in the dose maximum. The attenuation factor is the filter factor, defined as free dose rate/filtered dose rate. This factor is correlated to your isodose angle in the reference depth of 10 cm. If you enlarge your filter but if you want to preserve your angle, you need the same central ray attenuation factor, that means the same middle lead thickness and the same slope of your filter. You dont need another wedge.

So you see, I have problems to understand your task.

If you want help or comments, may be you could construct a small drawing with your perspective.

I fail to see the reason why you just don't stick with the flattening filter free (FFF) mode as the alternate to the flat beam. I feel it just sounds somewhat more complex than it needs to be, I think you would need to alter your beam current to have a thinner flattening filter otherwise I think you would just end up with a cone shaped distribution. Maybe Hanno can answer your question better............

Correct if I am wrong. What I understood was MU is measured using an Ion chamber kept below the flattening filter.

Now I have a LINAC which can deliver 300 MU/min with a flattening filter for a field of 40X40 at a max PRF. Now if I want to deliver a higher MU/min ( not an SRS or SBRT with FFF) can I redesign my flattening filter such that it spreads across a lesser field say 20 X 20 but deliver a higher MU/min like 600 MU/min. I just want to check the possibility of doing this.

My strength In physics is very poor, I am basically a mechanical engineer trying to understand the physics part of radiation, so my questions may not sound logical.

What I want to try is vary the slope of the filter and reduce the centre line thickness to increase the MU/min, and I will not be able to preserve my angle. Still I can get a flattened beam of small field size with higher MU/min. Can this be any advantage from the user point of view ??

If you have a think about your diagram the unflattened beam that is emerging from the target is always the same forward peaked x-ray beam. However where you collimate it with the jaws will affect how much of the forward peaked shape that is seen and consequently how much flattening is required on the central beam. If we look at say a 5x5cm FFF beam it "appears" flat or close enough not to worry but as the field size increases we "see" more and more of the forward peaked shape. An example of what you are thinking about would be the Novalis Tx linac which has a limited field (10x10cm) and does have a higher dose rate and do have a thinner FF than what is seen in a conventional linac.

Interestingly I came across this http://arxiv.org/pdf/1306.5051 which is approaching the FFF beam in a different manner to achieve flattness.

As for whether it will be useful.................. need to think about that one some more

Lindsay is right, I was mistaken and spoke about wedge filters and therefore I didn´t understand your question.

I will try it again. If you omit the flattening filter you get a gaussian dose profile with a width depending on primary energy. If you don´t use flattening filters for constant energy but different field sizes , you still will get the central part of the gaussian. The highest slope will be near the central ray, thats why I suppose that diminuishing the thickness of a flattening filter in the central region to improve the dose rate will not deliver flat profiles, which can be used for therapy except for very small field sizes.

The alternative is the divergent electron beam like Lindsay mentioned it or a scanning beam with a nearly gaussian shape which is moved across the field. This techniques has been used til now only for electron beams, where the electron gaussian was scanned. It could possibly work also for the elentron ray before hitting the bremstarget. But thats a matter of experiment or Monte Carlo calculations. Don´t forget, that the other materials around the primary ray contribute to the dose rate by scattering of the divergent beam.

Does the varaian which claims dose rate of 600 MU/min is across the full field of 40X40 or for a smaller fields. Can you clarify on this...

I try to understand this question. Dose rate as 600 MU is the monitor signal. Its independent from field size, because it´s measured above the collimator.

Measured in the phantom you get a modification of this information, because you find scatter contributions, which enlarge the dose rate in the phantom. Simultaneously you diminuish the back scattered radiation from the collimator leaves back into the monitor chamber. Both effects show the cited field size dependence of the dose rate. Quantitative results can be found in my chapter for photon dose distributions page 352, Fig. 16.13. The dose rate variation with field size is about 20%.

Very old post but I will take a crack at it.

Regarding multiple flattening filters for one energy:

Some Varian models have an "SRS" beam, which is a flattened 6MV beam that can deliver a 1000 MU/min dose rate. It accomplishes this high dose rate by placing a smaller, shorter, truncated flattening filter in the beam path. Because it is truncated, it will only treat a maximum field size of 15x15. Dmax is slightly more shallow than the standard 6MV beam (1.5cm vs 1.6cm) due to the softer energy spectrum. This beam uses the same target as the standard 6MV beam, but needs independent absolute dose calibration because of the higher photon fluence through the smaller flattening filter.

I hope this answers your original question.