The physics of TG51

1
The physics of TG-51
D.W.O. Rogers National Research Council of
Canada Ottawa AAPM Refresher course 730am
Monday July 23,2001 Salt Lake City
Meeting http//www.irs.inms.nrc.ca/inms/irs/irs.ht
ml
2
(No Transcript)
3
General formalism

• kQ is the quality conversion factor
• accounts for ND,w variation with Q

4
Overview - photons

• get a traceable
• measure dd(10) with lead foil
• deduce dd(10)x for open beam
• measure Mraw at 10 g/cm2
• M PionPTPPelecPpol Mraw

5
Overview -photons (cont)

• lookup kQ for your chamber

6
Overview - electrons

• get a traceable
• measure I50 to give R50
• deduce dref 0.6 R50 -0.1 cm
• measure Mraw at dref
• M PionPTPPelecPpol Mraw

7
Overview -electrons(cont)

• lookup kecal for your chamber
• determine (fig, formula)
• establish (Mraw 2 depths)

8
Fundamentals - definitions
Absorbed dose calibration factor
Beam quality conversion factor
9
Fundamentals - definitions
Electrons electron quality conversion factor
gradient correction factor
photon-electron conversion factor
10
Dose equations
photons
electrons
11
Absorbed-dose calibration factors
electrons
These can be used to measure kecal and
12
Relationships of calibration factors
13
Spencer-Attix cavity theory
Kh is humidity correctiongtair is dry M is fully
corrected charge
14
Pwall

• accounts for wall not being water
• unity for electrons (ignores known backscatter
  issues for plane-parallel chambers - 1-2
  effects)
• same as TG-21 for photon beams

15
Pwall 0.05g/cm2 walls
16
Mass energy absoption coefficients
TG-51 data based on Hubbell 1982 and Cunninghams
MC calns
17
Stopping power ratios
TG-51 uses stopping powers from ICRU Report 37
18
Prepl Pgr Pfl
Prepl replacement correction accounts for
changes caused by the cavity
Pgr gradient correction fluence moves
upstream because of low density
Pfl fluence correction other changes due to
cavity
19
Pgr gradient correction
-fn of gradient chamber radius -taken as 1.00
at dmax Two approaches -effective point of
measurement for depth-dose curves -Pgr
multiplicative correction for absolute
dose measurements
20
Effective point of measurement
Johansson et al (1977) electrons 0.5rcav upstream
of central axis photons 0.6rcav (was 0.75rcav
previously) Only for depth-dose curves
with cylindrical chambers
21
Pgr electron beams
-for cylindrical chambers
-equivalent to using the effective point of
measurement -but allows rigorous definition of
calibration factor
22
Pgr photon beams
-TG-51 uses the calns of Cunningham and Sontag
(1980) (as did TG-21) -there is considerable
variation in data on this correction
23
Pgr photon beams
TG-51 uses the ratio PflQ/PflCo gt
reduced uncertainty
24
Pfl fluence correction
photon beams -fluence corrections not needed
-due to transient CPE electron beams TG-51 uses
same factors as TG-21 for cylindrical chambers
and same factors as TG-39 for plane-parallel
25
Pfl cylindrical
New data agrees well with that used in TG-51 Need
as f(R50) at dref
Wittkamper et al PMB 38(1991)1639
26
Pfl plane-parallel
TG-51 uses data from TG-39 (with new fit)
TG-39 MP 21(1994)1251
27
Pfl Ez from R50
Harder relationship Ez Eo ( 1 - z/Rp) Eo 2.33
R50 21 beams used to fit Rp1.271R50-0.23
Ez 2.33R50(1-z/(1.271R50-0.23))
28
Pcel Al electrode correction
-for electrode of wall material any effect is
part of Pfl Ma and Nahum(93) showed aluminum
electrodes have an effect -larger in photon
beams -but biggest effect is in electron beams
because it cancels in photons
Ma Nahum PMB 38 (1993) 267
29
Pcel Al electrode correction
-expts confirm calns but not as precise
Palm Mattsson PMB 44 (1999) 1299
30
Equations for kQ,kecal
-defn of absorbed-dose calibration factor gives
-combining Dmed Dair eqns
31
Equation for kQ
-using defn of kQ assuming W/e constant gives
-applies to electrons and photons but for e-, Pgr
depends on users beam
32
kQ components
33
Equations for kecal
-from defns of kecal
34
Realistic electron beam sprs
Ding et al Med Phys 22 (1995)489
35
Effects of realistic sprs
Ding et al MP 22(1995)489
36
Realistic sprs dref0.6R50 - 0.1
Burns et al MP 23(1996)383
37
(No Transcript)
38
Summary so far

• Have reviewed
• the formalism
• the equations
• how each factor is obtained
• the effects of different data bases
• How good is it?

39
Measurement of photon kQ
Seuntjens et al at NRC measured kQ for gt3 of
each of 6 chamber types Measured against primary
standards Measurement accuracy 0.5 kQ
consistent for each type RMS deviation TG-51 vs
expt for 60 data points is 0.4
40
Measured kQ vs TG-51
Seuntjens et al, MP 27(2000)2763
41
Measured kQ vs TG-51
Seuntjens et al, MP 27(2000)2763
42
Measured kQ vs TG-51
Seuntjens et al, MP 27(2000)2763
43
Measured kQ vs TG-51
Seuntjens et al, MP 27(2000)2763
44
Measured kQ vs TG-51
Seuntjens et al, MP 27(2000)2763
45
What is uncertainty on kQ?
- uncertainty on calculated kQ various
people estimate of 1 -but how uncertain is the
user? given the good agreement with
measurements? (rms dev lt 0.5) -a reasonable
uncertainty on TG-51 kQ values is
0.5
46
What is uncertainty on dose?

• Uncertainties (photons)
• on ND,w is 0.5-0.6
• on kQ is 0.5
• on M (dd(10)x, monitor etc) 0.7
• total uncertainty 1.0

47
Odds and ends

• Pion
• new equations
• problems with the theory
• stopping power ratios for depth-dose curves
• need sprs for realistic beams

48
Pion equations
Pion continuous beams (as TG-21)
Pion pulsed or pulsed swept (Pionlt1.05)
49
Pion equations
TG-51 eqn is good to 0.2 or 0.4 for pulsed or
pulsed-swept Pion lt 1.05
50
sprs for depth-dose curves
TG-51 gives the dose at dref To get the dose at
dmax requires a high-quality depth-dose
curve Need to correct for spr and Pfl
(cylindrical chambers) Need realistic spr vs
depth to be consistent with spr at dref
51
L/?(R50,z)
Burns et al gave a fit to the Monte Carlo
realistic spr values
Tabulated vs R50 and z/R50 at http//www.irs.inms.
nrc.ca/inms/irs/papers/PIRS719/pirs719.html
52
How accurate is L/?(R50,z)?
Fit done for 0.02 lt z/R50 lt 1.1 rms deviation
0.4, max dev 1 -outside this range, the fit
deteriorates rapidly -at dref, the fit is within
0.1 for R50gt3 cm, 0.3 at R501.5cm
53
L/?(R50,z) vs dref formula
i.e. agreement is very good
54
L/?(R50,z) vs original data
agreement not so good 0.6 variations near
dmax failsgt1.1R50
55
Summary
There is too much in TG-51 to cover in 1
lecture I emphasized fundamentals and how
quantities are calculated I didnt even mention
beam quality specification!
56
Resources
A formal review paper is coming! The slides from
this and related TG-51 talks are available at
http//www.irs.inms.nrc.ca/inms/irs/irs.html My
1996 Summer school lecture and 1998 paper (MP 25
310-320) cover many details about TG-51 (on web)