dosimetry for imrt

Dosimetry for IMRTDosimetry for IMRT

Thomas Rockwell Mackie

Department of Medical PhysicsUniversity of Wisconsiny

Madison [email protected]

Financial Disclosure

I am a founder and Chairman of TomoTherapy Inc. (Madison, WI) which is participating in the ( , ) p p gcommercial development of helical tomotherapy.

Outline

• Dosimetry Issues Relevant to IMRT• Charged Particle Equilibrium• Temporal Non-Constancy• Partial Volume EffectsPartial Volume Effects

• Non-Standard Fields • IMRT as a Large Collection of Small Non-Standard Fields• IAEA Calibration Initiative for Non Standard Fields• IAEA Calibration Initiative for Non-Standard Fields

• Patient-Specific Methods To Verify Delivery of IMRT• Independent Calculations • Delivery of Individual Radiation Fields• Delivery of All Fields into a Phantom • Metrics for Patient-Specific Dosimetry QAMetrics for Patient Specific Dosimetry QA

Dosimetry Issues Relevant to IMRT

• Charged particle equilibrium– Different spectrum for collection of small fields

N if d– Non-uniform dose

• Temporal non-constancyA very small effect for ion chambers– A very small effect for ion chambers

– May not be true for other dosimeters

• Partial volume effect• Partial volume effect– Most important effect especially when measuring

output factors for small fieldsp

Non-Uniform Intensity Changes the Energy Spectrumgy p

More High EnergygyElectrons

Photon Photon

More Low EnergyElectronsPhoton

BeamletPhotonBeamlet

Electrons

Change in Stopping Power RatioComparison of Spencer-Attix (∆=10 keV) restricted mass collisional stopping powers ratios (water/air) at 5 cm depth in water for various 6 MV beams with stereotactic and MLC beams.

6 MV beams Beam quality, TPR(20,10)

Andreo, (1994)

Sánchez-Doblado, (2003)

Column4/Column3

El kt SL 18 0 690 1 1187 1 1188 1 000Elekta SL-1810 x 10 cm2

0.690 1.1187 1.1188 1.000

1.0 cm diameter stereo field

1.1155 0.997

0.3 cm diameter stereo field

1.1153 0.997

Siemens Primus10 10 2

0.677 1.1213 1.1221 1.00110 x 10 cm2

2 x 2 cm2 irregularon-axis beamlet

1.1203 0.999

2 x 2 cm2 irregular8 cm off-axis

1.1250 1.003

Temporal Delivery of IMRT

IMRT Delivery Signatures

Delivery of Dose to a Single Voxel

12000

Step & Shoot HIART

8000

10000

4000

6000

Sign

al (f

C)

0

2000

0.16

46.9

93.7

141

188

235

282

329

376

423

470

517

564

613

658

702

747

792

836

881

926

971

015

060

105

149

194

239

-2000

0 46 93 1 1 2 2 3 3 4 4 5 5 6 6 7 7 7 8 8 9 9 10 10 11 11 11 12

Elapsed Time (sec)

From Tim Holmes, St. Agnes Baltimore

Temporal Delivery of IMRTD li f D t Si l V l

Step and Shoot Helical Tomo

Delivery of Dose to a Single Voxel

0.80.9

1

ve D

ose

0 40.50.60.7

Cum

ulat

i

0 10.20.30.4

rmal

ized

5 %

90 % 5 %

00.1

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

No

Elapsed Time (minutes)

From Tim Holmes, St. Agnes Baltimore

Partial Volume Effect

Dose to Water For Small Fields

0

0.8

0.9

r M t C l

0 5

0.6

0.7

ut F

acto

r Monte CarloDiamondDiodeLAC+filmFilm

0 3

0.4

0.5

Out

pu

PTW 31014 (Pinpoint)PTW 31010 (Semiflex)PTW 30006 (Farmer)

0.5 1.0 1.5 2.0 2.5 3.00.2

0.3

Side of Square Field (cm)Side of Square Field (cm)

From Roberto Capote, IAEA

Partial Volume Effect

output factors relative to 4 cm x 4 cm

1 00

1.10

0.80

0.90

1.00

r rel

to 4

x4

0.60

0.70

outp

ut fa

cto

standard dataphoton diodedi d

0.40

0.50

o diamondpinpoint0.125 cc chamberelectron diode

0 1 2 3 4 5 6 7 8 9 10equivalent field lenght (cm)

From Karen Rosser, Royal Marsden

High Uncertainty in Output FactorsE l St ti ti f 45 O t t F t f 6 d 18 fi ldExample: Statistics of 45 Output Factors for 6 mm and 18 mm square fields

(Novalis, SSD = 1000 mm, depth = 50 mm, various detectors)

Factor of Two in Beam Calibration!

From Wolfgang Ullrich, BrainLab

Reasons for Drop in Output with Small Field Size

• Backscatter into monitor unit from beam defining jaws

• Reduced scatter (phantom and head)• Reduced scatter (phantom and head)• Electronic disequilibrium• Obscuration of the source

Problems with Conventional Output Factors

• Small field openings obscure

Factors

p gthe source which is difficult to measure and error prone.

• Amount of phantom scatter changes as well as lateral disequilibriumdisequilibrium.

• Partial volume effects can mask machine outputmachine output factor.

Standard Reference Conditions

Small Field Conditions

Scan Phantom Across a Narrow BeamIf the source were a point source the• If the source were a point source, the measured primary dose would be directly proportional to the jaw field width and inversely proportional to thewidth and inversely proportional to the scan speed.

• Nearly equal phantom scatter conditions for all jaw settings.for all jaw settings.

• No partial volume effects once scanning is complete.

Beam’s eye viewJaw field width (FW)

Animation of Scanning PathFrom the Point of View of the PhantomFrom the Point of View of the Phantom

FW i the f be idthThe fan beam

Th f b

FW is the fan beam widthThe fan beam field 10 cm long by FW wide b i The fan beam

field is always scanned the

begins at one side of the field and is scanned

same distance (10 cm) at the same speed

across the phantom. FW is defined by the

Beam’s eye view

same speed. defined by the 50% to 50% dose.

Scanning at Velocity vAssume that the source is a point source

Jaw field width (FW)

Assume that the source is a point source

r( )

D r( )( )

10 cm

Animation of Scanning PathFrom the Point of View of the PhantomFrom the Point of View of the Phantom

FW i the f be idthThe fan beam

Th f b

FW is the fan beam widthThe fan beam field 10 cm long by FW wide b i The fan beam

field is always scanned the

begins at one side of the field and is scanned

same distance (10 cm) at the same speed

across the phantom. FW is defined by the

Beam’s eye view

same speed. defined by the 50% to 50% dose.

Scanning at Velocity v

Jaw field width (FW)

r( )

FWD

D r( )

Dv

D r( )

10 cm

Impact of Source Occlusion for Variable Jaw WidthsJaw Widths

Dashed line is t dWater Tank Dose Ratios

16

18

20

expected result for a point source

8

10

12

14

Dose Ra

tio

• Only the field width, FW for each se

2

4

6

8D each measurement was altered.

• The scan distance

Dos

0

0 1 2 3 4 5

Measured Field Width, FW (cm)

(10 cm) and scan velocity were identical.Measured Field Width, FW (cm)

Mainly Due to Source Obscuration

Dose and Dose/FW as a Function of Field Width (FW)Function of Field Width (FW)

Divide by Field Width and Normalize

Water Tank Dose/FW Ratios

1.2

Water Tank Dose Ratios

20

0.8

1

W Ratio

Basically a measure of the clipping the photon so rce seen

12

14

16

18

atio

ose/

FW0.2

0.4

0.6

Dose/FW photon source seen

for the smaller field widths.

2

4

6

8

10

Dose R

Dos

e

orm

aliz

ed D

o

0

0 1 2 3 4 5

Measured Field Width, FW (cm)

Plot of dose D(FW) Plot of D/FW

0

2

0 1 2 3 4 5

Measured Field Width, FW (cm)Measured Field Width, FW (cm) Measured Field Width, FW (cm)

N

Plot of dose D(FW) Plot of D/FW

From Brian Hundertmark

No Difference Between Water and Solid Water


Little Difference With Depth


Small Vs. Farmer Chambers

A1SL: 0 057 ccA1SL: 0.057 ccPTW: 0.6 cc


Helical Delivery


The Same Technique Could Be Applied for Other Small BeamsOther Small Beams

Now for a point source, the measured primary dose would be proportional to the area of the beam andwould be proportional to the area of the beam and inversely proportional to the scan velocity.

You wouldA circular beam is scanned in a You would

actually scan the phantom instead of

raster pattern to create a broad beam using the

Beam’s eye view

instead of scanning the beam.

beam using the same overall irradiation time.

Avoids Problems with Measuring Output Factors forMeasuring Output Factors for

Small Fields

• Charged particle equilibriumg p q– Nearly same charged particle spectrum

• Temporal non-constancyTemporal non constancy– Deliberately uses temporal non-constancy– Mimics how composite small fields become– Mimics how composite small fields become

larger irradiated fields

• Partial volume effect• Partial volume effect– No partial volume effects

Measurements Would Benchmark Source Model

Model the source distribution

Calculate a delivery to a large area using a

superposition of small bbeams

Measure the dose in the center of the fieldcenter of the field

Repeat for another set of ll b til fsmall beams until a range of beam sizes are done.

Agreement

no

Enough? between calculation and measurement?yes yes

Doneno

0.05 0.30 0.55 0.80

IMRT Is Far from Reference Conditions0.05 0.30 0.55 0.80

0.10 0.35 0.60 0.85

0.15 0.40 0.65 0.90

0.20 0.45 0.70 0.95

0.25 0.50 0.75 1.00

From Art Boyer, Stanford

Audit for TRS 398 Reference Dosimetry

10

468

on (%

) Farmer Semiflex PinPoint

024

Dev

iatio

-6-4-2

ativ

e D

-10-86

Rel

CASE NUMBER1 2 3 4 5 6 7

CASE NUMBER

From Roberto Capote, IAEA

IMRT Audit

810

Farmer Semiflex PinPoint

%)

2468

DynamicStep-and-shoot

atio

n (%

-202

e de

via

-6-42

Rel

ativ

e

1 2 3 4 5 6 7 8 9 10 11 12 13

-10-8R

CASE NUMBER1 2 3 4 5 6 7 8 9 10 11 12 13

Sánchez-Doblado F, Hartmann G., Pena J., Capote R. et alIJROBP 68 (2007) 301-310

IMRT Is About Using Small FieldsIMRT Is About Using Small FieldsIntensity pattern

possibly unconstrained p yintensity levels

What is delivered

Intensity limited to a few discrete intensity levelsy

Adapted from Michael Sharpe, U. of Toronto

Leaf Penumbra is Important

Lines DefiningOne Half Value Layer of Beam

Attenuation

Lines DefiningLight Field Boundary

CentralAxis

Boundary

• It is important to have an accurate model of the curved leaf endsthe curved leaf ends.

• More important for summation of small fields.

Meeting for the Dosimetry Code of Practice:Meeting for the Dosimetry Code of Practice:Small Fields and Novel Beams 3-5/12/07

• Jan Seuntjens

IAEA• Pedro Andreo

Outside Participantsj

• Hugo Palmans• Karen Rosser

• Ken R. Shortt• Stanislav VatnitskiyKaren Rosser

• Saiful Huq• Wolfgang Ullrich

y• Roberto Capote• Joanna Izewska• Wolfgang Ullrich

• Warren Kilby• Rock Mackie

Joa a e s a• Ahmed Meghzifene• Rodolfo Alfonso• Rock Mackie Rodolfo Alfonso

• My Pictures\IAEA_2008\IAEA_2008 006.jpgjpg

Examples of Small and Novel Fields

• GammaKnife (1 8 cm diameter collimator is the• GammaKnife (1.8 cm diameter collimator is the largest collimator – intrinsically composite field –not 100 cm SSD)

• Linac SRS beams (extrapolate to small field conditions)

(6 di lli i h l• Accuray (6 cm diameter collimator is the largest collimator)

• TomoTherapy (5 cm is the largest slice width• TomoTherapy (5 cm is the largest slice width –no field flattening filter)

• IMRT (made up of numerous small fields)IMRT (made up of numerous small fields)

Static Field CalibrationUses a machine-specific reference field, fmsr

msr

msr

Q Cow D ,w Q QD M N k k 60

msrQ( D / M )k

msrQDfmsr

msr

wQ Q

w

( D / M )k

( D / M )wD

Corrects for the differences between the conditions of field size, geometry, and beam quality of the conventional reference field g y, q yand the machine-specific reference field.

Calculate Using MCUsing method of Sempau et al 2004 PMB 49;4427-44

msrQw aD D

k( / )

Using method of Sempau et al 2004 PMB 49;4427-44

msrQ Qw a

kD D( / )

is the dose at the reference point in water

D

wD

is the average dose in the ion chamberaDaD

DwD

Adapted from Edmond Sterpin

Composite Field CalibrationpUses a plan-class specific reference field, fpcsr

pcsr

pcsr

Q Cow D ,w Q QD M N k k 60

pcsrQ( D / M ) fpcsr

pcsr

wQ Q

w

( D / M )k( D / M )w

Corrects for the differences between the conditions of field size, t d b lit f th ti l f fi ldgeometry, and beam quality of the conventional reference field

and the plan-class specific reference field.

Dose Calibration Correction Factor to G t D i Cli i l Fi ldGet Dose in a Clinical Field

M clinclin msr clin msr

msr clin

QQ Q Q Qw w Q w Q

Q

MD D D k

MmsrQM

clinQ( D / M )

clin msr

wQ Q

w

( D / M )k

( D / M )

In principle, the dose calibration correction factor would be different for each clinical measurement. Eventually, Monte Carlo treatment planning systems could compute for both the patient and a QA phantom.

clinQk

Overview of Static Field Dosimetry

REFERENCE DOSIMETRY

msr

msr

Q Cow D ,w Q QD M N k k 60 clin m sr clin

m sr

Q Q Qw w QD D

10 cm x 10 cm Reference Field

msrQk

Linac Stereotactic Field

CoD ,w QN k 60

BrainLab MiniMLC (10 cm x 10 cm)

Hypothetical 10 cm x 10 cmReference Field clin

m sr

QQ

CyberKnife (6 cm Diameter)

Qk

Reference Field

msrQ

TomoTherapy(5 cm x 10 cm )

Ion Chamber =

“Effective kQ” for Tomo

1. Determine the %dd(10)x[HT Ref] for a tomotherapy unit for a 5 cm x 10 ( ) [ ] pycm field at 85 cm SSD.

2. Using the graph at the right look up the value of %dd(10)x[HT TG-51]. 3. Using the graph on the left and the derived value %dd(10)x[HT TG-51]

determine the abscissa and this effective kQ value is then called in the IAEA protocol.

msrQ Qk k

Thomas et al (2005)

Static and Composite Field C l l ti f TCalculations for Tomo

Static Field Calibration Composite Field CalibrationStatic Field Calibration(Section III.A.1)

Composite Field Calibration(Section III.A.2)

QkkJeraj et al 2005 Duane et al 2006

5 cm x 10 cm 0.997 Unmodulated Helical Delivery5 cm Slice Width

1.000pcsrQk

msrQk

2 cm x 10 cm 0.993 Unmodulated Helical Delivery2.5 cm Slice Width

1.000

2 cm x 2 cm 0.990 Unmodulated Helical Delivery 0.9971 cm Slice Width

ICRU Committee on IMRTSponsors

• André Wambersie MD, PhD, Brussels, Belgium• Paul DeLuca PhD, Madison, USA

R i h d G hb MD Ph D Cl l d USA• Reinhard Gahbaue MD, Ph.D, Cleveland, USA• Gordon Whitmore Ph.D, Toronto, Canada

ChairmenVi t G é i MD PhD B l B l i• Vincent Grégoire MD PhD, Brussels, Belgium

• T. Rock Mackie PhD, Madison, USA

MembersWilfried De Neve MD PhD Gent Belgium• Wilfried De Neve MD PhD, Gent, Belgium

• Mary Gospodarowicz MD, Toronto, Canada• Andrzej Niemierko PhD, Boston, USA• James A. Purdy PhD, Davis, USAy• Marcel van Herk PhD, Amsterdam, The Netherla• Nilendu Gupta PhD, Colombus, USA

Consultants• Anders Ahnesjö PhD, Uppsala, Sweden• Michael Goiten PhD, Windisch, Switzerland

Supplement to ICRU-50 and ICRU-62

• More emphasis on statistics.

• Prescription and reporting with dose-volume specifications.p

• No longer use ICRU-Reference Point.

• Want median dose D50% reported.

• Use model-based dose calculations.

• Include the effect of tissue heterogeneities.

R t d t ll f t t d• Report dose to small mass of water, not dose to tissue.

QA for IMRTQ f S• Appropriate QA of TPS and delivery

equipment

• Patient-specific QA:• Delivery of individual fields into a• Delivery of individual fields into a

dosimeter

D li f ll f th fi ld i t h t• Delivery of all of the fields into a phantom

• Independent dose calculation algorithms with similar of better dose calculation accuracy

• In-vivo dosimetry not limited to a single point.

G DGap Error

Gap error Dose error20.0

15.0

rror

Range of gap width

10.0

Dos

e er

2.0

Gap error (mm)

5.0% D 1.0

0.50.2

0.00 1 2 3 4 5

Nominal gap (cm) g p ( )

From Tom Losasso, Memorial Sloan Kettering

Errors Introduced

Leaf Positioning Errors1 mm Bands

- 0.5 mm

- 0.2 mm

+ 0.2 mm

+ 0.5 mm

Mechanical Alignment of gStatic MLC Approach

No offset 0.6 mm offset 1.0 mm offset

From Jatinder Palta, University of Florida

Radiation Field Edge Offset

0.7mm offset 0.6mm offset

0.3mm offset0.5mm offset

Segmental Multileaf Collimator (SMLC) D li S t(SMLC) Delivery System

T l A tiTolerance Limit

Action Level

MLC*Leaf position accuracyLeaf position reproducibilityGap width reproducibility

1 mm0.2 mm0 2 mm

2 mm0.5 mm0 5 mmGap width reproducibility 0.2 mm 0.5 mm

Gantry, MLC, and Table 0.75 mm di

1.00 mm di

* Measured at all four cardinal gantry angles

Isocenter radius radius

Beam StabilityLow MU Output (<2MU) 2% 3%Low MU Output (<2MU)Low MU Symmetry (<2MU)

2%2%

3%3%

Jatinder Palta, AAPM Summer School 2003

Dynamic Multileaf Collimator (DMLC) Delivery System(DMLC) Delivery System

Tolerance Li it

Action L lLimit Level

MLCLeaf position accuracy 0 5 1Leaf position accuracyLeaf position reproducibilityGap width reproducibility

0.5 mm0.2 mm0.2 mm

1 mm0.5 mm0.5 mm

Leaf speed 0.1 mm/s 0.2 mm/s

Gantry, MLC, and Table Isocenter

0.75 mm radius

1.00 mm radiusIsocenter radius radius

Beam StabilityLow MU Output (<2MU) 3% 5%p ( )Low MU Symmetry (<2MU)

3%2%

5%3%

Jatinder Palta, AAPM Summer School 2003

Fail FailFail Pass

Leaf open time histograms (15 sec gantry)Higher Avg. Leaf

Opening Time

Leaf UsageLeaf opening time

Leaf #

TomoTherapy Leaf Latencypy y

• TomoTherapy uses pylinear fit of measured data to model leaf latencmodel leaf latency

• Based on small lead opening times lead opening times we expect leaf latency as the cause

replannedoriginal plan

pitch = 0.143 pitch = 0.287

patient 1

PTV 70PTV 64PTV 54PTV 50larynxbrainstemr. parotid

pitch = 0.287pitch = 0.215

patient 2

PTV 54esophagusp glungscordl. ventricle

patient 3

pitch = 0.287pitch = 0.215PTV 60PTV 50esophaguslarynxcordr. parotidb i tbrainstem

End-to-End QA

QA Measurements

Measure “Cheese”Phantom

d f Q

plane and point dose at the sameused for QA

measurements

at the same time

Film PlanePhantom can be rotatedPhantom can be rotated or turned to acquire any

orthogonal plane

On and Off-Axis Results

Delivered Dose: 2 5cm Treatment Beam

Film and Ion Chamber Absolute Dose

Delivered Dose: 2.5cm Treatment Beam

2.5

On-Axis Tumor Off-Axis Tumor

1.5

2.0

Gy)

1.0Dos

e (G

0.0

0.5

-20 -15 -10 -5 0 5 10 15 20

Distance (cm)

QA for AllQA for All of the Fields

TomotherapypyExample

Delivery QA Panel

Comparison of Phantom Plan and Verification Film

Plan

FilmN t thNote theHighGradients

Film

Plan

From Chet Ramsey, Thompson Cancer Survival Center

QA of Individual FieldsQ

External diode/ion-chamber arrays• MapCheck• PTW Octavius phantom• IBA Matrix• IBA Matrix

Integrated detector systems• EPID portal dosimetryp y

Independent Calculation

Dose Accuracy and Distance to AgreementAgreement

Dose accuracy for low gradients Dose

m c m c 50r r D r D r D %( , ) ( ) ( ) /

Dose

Low gradients < 20%/cmMeasured

m c m cr r r r r( , )

Distance to agreement for high gradients

ComputedComputedRadius

Gamma Function2 2 1 2

m c m c M m c Mr r r r D r r r d /( , ) {[ ( , ) / ] [ ( , ) / ] }

MDDose accuracy axis

m c m c M m c M

m cr r( , )

M

Passm c

Distance to agreement

m cr r r( , )

d

agreement axis

Md

Gamma Function2 2 1 2

m c m c M m c Mr r r r D r r r d /( , ) {[ ( , ) / ] [ ( , ) / ] } m c m c M m c M

MDDose accuracy axis

m cr r( , )

M

Failm c

Distance to agreement

m cr r r( , )

d

agreement axis

Md

IMRT Evaluation using Anthropomorphic PhantomsAnthropomorphic Phantoms

For H&N, using a criteria of 5% or 4mm, the i t d f 75% t 58%Molineu et al IJROBP 2005

Ibott et al Tech in Ca RT 2006Followill et al Med Phys 2007

Courtesy Ibott, RPC

passing rate drops from 75% to 58%

Gortec IMRT Test PhantomTLDs are placed at seven locations.

Point 1: Isocenter

P i t 2 S i l d i t Point 2: Spinal cord isocenter

Point 3: Spinal cord cranial

Point 4: PTV T R

Point 5: PTV T R cranial

Point 6: PTV N L

Point 7: PTV N L caudalCourtesy M. Tomsej,Brussels

Audit Results

Dm/Dc=f(CENTER) per meas. pt

1.15

1.2

1.05

1.1

isocenter

spinal cord iso

i l d i l

0.95

1

Dm/D

c spinal cord cranial

PTV T D

PTV T D cranial

PTV N G

PTV N G caudal

0.85

0.9

Sample Result

0.80 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

CENTER

Sample Tomotherapy Results

Inter-Institution Dose AccuracyAccuracy Distribution

600

300400

500

uenc

y

0100

200Freq

00.885 0.905 0.925 0.945 0.965 0.985 1.005 1.025 1.045 1.065 1.085 1.105 1.125

Measured Dose/Computed Dose

Number of Measurements = 2679Mean = 0.995Standard Deviation = 0 025Standard Deviation = 0.025

(Updated from Zefkili et al 2004)

Intra-Institution Dose AccuracyAccuracy Distribution

300

350

200

250

300

uenc

y

50

100

150

Freq

0-10 -8 -6 -4 -2 0 2 4 6 8 10

Relative Difference Between Measured and Calculated Dose (%)

Number of Measurements = 1591Mean = 0.45%Standard Deviation = 2.5%

(Updated from Dong et al 2003)

QA Accuracy for IMRTy• Previous ICRU 5% point-dose accuracy

ifi ti l d b l t i dspecification replaced by a volumetric dose accuracy specification.

• Proposed new ICRU volumetric dose accuracy specification:

- High gradient (≥ 20%/cm): 85% of points within 5 mm (3.5 mm SD),

- Low gradient (< 20%/cm): 85% of points within 5% of predicted dose normalized to the prescribed dose (3.5% SD).

Monitor Units Calculations for M d l B d D C l l tiModel-Based Dose Calculation

D DA r A r A r b AMU MU

10

0

( , ) ( , ) ( , ) (1 ( ))

0

ComputedDirectly

Beam Output

Correction toAccount forBackscatterBackscatterInto MonitorChamberP

d A

Monitor Units Calculations for M d l B d D C l l tiModel-Based Dose Calculation

D cal cal

Measuredcal

D A dM b A

MU D0

( , )(1 ( ))

dcal Acal cal

Calculated

U D A d0

( , ) Ref

Not including the effect of backscatter into the it h b ill lt i b t 2% tmonitor chamber will result in about a 2% error at

worst.

Backscatter into Monitor ChamberThe effect is due to backscattered photonsThe effect is due to backscattered photons entering the monitor and resulting in feedback to the linac to lower its output

Varian 2100 – 10 MV. Results with other jaw completely open

Liu et al., Med. Phys 2000;27:737-744

Monitor Backscatter for Square On Axis FieldsOn-Axis FieldsVarian 2100 – 10 MV

Liu et al., Med. Phys 2000;27:737-744

Conclusion• IMRT uses complex field boundaries and many small

fields to achieve more conformal dose distributions. • Partial volume effects can result in severe error

especially when measuring the dose in high gradients.• A new method for determining the effect of field

obscuring is proposed to eliminate partial volume g p p peffects.

• IMRT deliveries are far from the measurement conditions of calibration.

• IAEA has developed a formalism to account for small and novel beams.

• Independent calculations or measurements for patient-Independent calculations or measurements for patientspecific QA are required for IMRT.

• IAEA will recommend that 85% of measurement values (dose or distance to agreement) are within 5%.(dose or distance to agreement) are within 5%.

• Model-based monitor unit calculations are simple but should include scatter into monitor chamber.

dosimetry for imrt

Documents