Page 1 of 13
`
`Chang & Gibbons 1999 AAPM RC
`Clinical Implementation of Non-Physical Wedges
`
`1999 AAPM Refresher Course
`
`Sha X. Chang, Ph.D.1 and John P. Gibbons, Ph.D.2
`1Department of Radiation Oncology, UNC School of Medicine, Chapel Hill, NC
`2Department of Medical Physics, Palmetto-Richland Memorial Hospital, Columbia, SC
`
`I. INTRODUCTION
`
`A. Definition.
`
`A non-physical wedge generates a spatial dose distribution similar to that produced by a
`physical wedge without a physical filter in the photon beam. Instead, an exponential fluence
`profile is produced via motion of one of the collimating jaws. Proposed in the late 1970s, non-
`physical wedges have been implemented on both Varian and Siemens’ accelerators as the Varian
`Dynamic Wedge (DW) and Siemens Virtual Wedge (VW). Recently, Varian has introduced the
`Enhanced Dynamic Wedge (EDW) to add functionality to this modality.
`
` B. Comparison of modalities
`
`Although similar in function, the Varian and Siemens implementation of non-physical
`wedges differ in many ways that users should be aware of. Table 1 highlights some of these
`differences:
`
`Table 1
`
`Feature
`Jaw Position vs MU
`
`Method of delivery
`
`Initial/Final Jaw Positions
`
`Wedge direction option
`
`Jaw travel limitations
`
`Enhanced Dynamic Wedge
`Determined using
`segmented treatment table
`(STT)
`Variation of dose rate and
`moving jaw speed
`
`Initially open; final position
`0.5 cm from fixed jaw
`EDW for Y (upper) jaws
`only. Treatment prohibited
`if fixed jaw >0.5cm beyond
`moving jaw limits
`
`Virtual Wedge
`Determined using analytic
`equation
`
`Variation of dose rate only
`
`Initially 1.0 cm from fixed
`jaw; final position fully
`opened.
`VW for X or Y jaws.
`Treatment allowed if fixed
`jaw >1cm beyond moving
`jaw limits
`
` Gradient direction
`
`10 cm pass CAX.
`
` Non-gradient direction
`
`No limit.
`
`Monitor Unit Input
`
`MUs = Total MUs delivered
`during treatment
`
`upper jaw: 2 cm pass CAX.
`lower jaw: 10 cm pass CAX.
`
`No limit.
`Programmed MUs = MUs
`delivered with CAX in the
`field. Total MUs termed
`MUmax.
`
`Page 1 of 13
`
`Elekta Exhibit 1032
`
`

`
`Page 2 of 13
`
`Chang & Gibbons 1999 AAPM RC
`
`Feature
`Wedge Angle Selection
`
`Wedge Factors
`
`Machine-independence
`
`Enhanced Dynamic Wedge
`7 wedge angles (10o, 15o,
`20o, 25o, 30o, 45o, 60o)
`
`Strong function of both
`wedge angle and field size;
`Weak function of off-axis
`distance.
`STTs same for all Varian
`machines
`
`Virtual Wedge
`Continuous to 60o; Larger
`angles available with
`reduced field sizes.
`Approximately unity (!5%)
`for symmetric fields; Strong
`function of off-axis distance.
`
`VW equation may vary with
`user-adjustable calibration
`factor c.
`
`II.
`
`MONITOR UNIT CALCULATIONS: NON-PHYSICAL WEDGE FACTORS
`
`A. Field Size Dependence
`
`Both DW and EDW show strong field size dependence. Measured DW factors (Fig. 1)
`exhibit a discontinuity between 9.5 and 10 cm width due to change in STT step size. Measured
`EDW factors (Fig. 2) are derived from a single table and have a smooth field size dependence.
`In both cases, the wedge factors have been shown to be closely approximated by the fraction of
`monitor units delivered with the central axis in the field (“MU fraction” model). Additionally,
`EDW factors appear to be machine-independent to within 1%.
`
`Figure 1
`
`Figure 2
`
`
`EDW factors can be determined in several ways. In addition to direct measurement,
`inspection of the STT prior to treatment to determine the MU fraction will provide adequate
`
`Page 2 of 13
`
`

`
`Page 3 of 13
`
`Chang & Gibbons 1999 AAPM RC
`
`prediction of the EDW factor in most cases. Since the EDW treatment STT is smooth, an
`analytic function6 may also be used to describe this quantity.
`Virtual wedge factors are close to unity for all symmetric fields of different wedge angles.
`The wedge factor will deviate from unity for asymmetric fields when the wedge factor is
`calculated on the center of the asymmetric fields. Measured virtual wedge factors (Fig. 3) show
`variation less than !5% for range of field sizes and wedge angles. The systematic deviation of
`VWF at large wedge angles and field sizes can be corrected using a wedge factor file/table for
`the VWF calculation.
`
`Figure 3. Virtual wedge factor vs. wedge angle
`
`6X, obs
`18X, obs
`6X,10x10
`18X, 10x10
`
`20
`
`40
`ANGLE (DEG)
`
`60
`
`80
`
`1.04
`
`1.03
`
`1.02
`
`1.01
`
`1
`
`0.99
`
`0
`
`VWF
`
`Variation from the “MU fraction” model may exceed clinical tolerance for MU
`calculations for large field size, wedge angle combinations. Measured values for these cases can
`be input into clinical tables. An extension to the “MU fraction” model can be used to determine
`both EDW and VW factors to within 2%. Resulting EDW factors using this approach for 6X and
`18X EDW factors are displayed in Tables 2 (left) and 3 (right), respectively.
`
`Page 3 of 13
`
`

`
`Page 4 of 13
`
`B. Depth Dependence
`
`Chang & Gibbons 1999 AAPM RC
`
`Unlike physical wedges, the dosimetly of non-physical wedges shows far less variation with
`depth in the absence of the beam hardening effect. A slight increase in measured PDD has been
`demonstrated with both Dynamic and Virtual wedges and this has been attributed to a secondary
`effect of the exponential fluence distribution. In most cases, the dosimetric variations are less
`than 2%.
`
`C. Off-Axis Dependence
`
`Both EDW and VW allow asymmetric fields in either the non-gradient and/or gradient
`directions. In the non-gradient direction, no deviation from open field Values has been reported.
`In the gradient direction, EDW factors can vary by up to 15%, while VW factors may vary by
`more than 100%. For wedge factors defined at the geometric center of the field, analytic models
`have shown agreement within 2%. Figure 4 and 5 displays off-axis EDW and VW factors for
`30° non-physical wedges for 6MV and l8MV photons, respectively.
`
`1.00
`
`o_g5
`
`0'90
`E 0.85
`E}
`0.80
`0.75
`
`0.70
`
`05520
`
`-10
`
`0
`
`10
`
`Field Center (off-axis)
`
`Figure 4. EDW factor of 30-
`degree wedge and 6MV.
`Three sets of data are for field
`size 5x5 (circles), 10x10
`(squares), and 20x20 (triangles).
`
`Figure 5. VW factor for
`6MV (squares) and 18 MV
`(circles) at field size of 5x5
`at different field center
`
`position.
`
`-20
`
`-15
`
`-10
`
`-5
`
`0
`
`5
`
`10
`
`OFF AXIS DISTANCE
`
`Page 4 of 13
`
`

`
`Page 5 of 13
`
`Chang & Gibbons 1999 AAPM RC
`
`III.
`
`Implementation of non-physical wedges into a Treatment Planning System (TPS)
`
`The dynamic nature of the non-physical wedge functions offers great ease for photon beam
`treatment delivery. However, it poses a considerable challenge to most of the available TPS',
`whose data structures inherently assume that the radiation beams are static. There are three types
`of implementation methods for incorporating a non-physical wedge of choice into a TPS:
`1) photon fluence modeling
`2) physical wedge emulation
`3) synthesis of two or more wedge fields
`Depending on the flexibility of the TPS on hand one can use at least one of the three types of the
`methods to incorporate the non-physical wedge function.
`
`1. Photon fluence modeling
`Photon fluence modeling is the ideal choice of all TPS implementation methods. It requires a
`sophisticated TPS that is able to model the photon fluence actually generated by the non-physical
`wedge delivery process. As a result, all aspects of treatment planning can be accurately
`performed with the consideration of all specific limitations of the non-physical wedge of the
`concern. Unfortunately, only a few TPS are truly equipped with such flexibility. This type of TPS
`includes the Univ. of North Carolina in-house TPS PLUNC [Chang et al 1999] and ADAC's
`Pinnacle3 [Bayouth & Steinberg] for VW. For EDW such TPS' are CadPlan of Varian-Dosetek
`[Salk et al, Samuelsson et al 1997], IsiS3D of Technologie of Diffusion [Papathoedorou et al
`1999], and Multidata DSS v2.35 [Beavis et al 1996]. Helax TMS TPS is also reported to have
`such a function [Karlsson 1997]. PLUNC computes the doses based on the photon fluence
`generated by the virtual wedge with the consideration of head scatter variation during the dynamic
`treatment delivery. Others TPS' model the photon fluence by superpositioning many segment
`treatment fields based on the STT table for EDW and the output rate analytical equation for VW.
`These methods realistically simulate the actual wedge treatment delivery and therefore produce
`reliable results in terms of both the relative dose distribution and the absolute MU calculation.
`The figure below by Bayouth & Steinberg shows there is a very good agreement between
`calculated and measured beam profiles of different wedge angles (Figure 6) of VW. Table 4
`displays the excellent agreement between measured wedge factors and the calculation by PLUNC
`for 6MV photon for both symmetric and asymmetric fields and at different depth.
`
`Figure 6. Beam profile data from
`Bayouth & Steinberg (unpublished).
`VW angles measured ranging from 10
`to 70 degrees. Excellent agreement
`between the measured data (solid lines)
`using linear array detectors and the
`corresponding calculation (symbols)
`using Pinnacle3 TPS from ADAC.
`
`Page 5 of 13
`
`

`
`Page 6 of 13
`
`Chang & Gibbons 1999 AAPM RC
`
`Table 4. 6 MV VW factor comparison (Univ. of North Carolina)
`
`Wedge dir. 15W
`
`30W
`
`45W
`
`60W
`
`1(calc.)
`1(meas.)
`
`1 (calc.)
`1 (meas.)
`
`1 (calc.)
`1 (meas.)
`
`1 (calc.)
`1 (meas.)
`
`1 (calc.)
`1 (meas.)
`
`0.997
`0.997
`0.0
`0.997
`0.997
`0.0
`1.0
`0.998
`0.2
`1.0
`0.999
`0.1
`0.926
`0.930
`0.4
`0.930
`0.933
`0.3
`1.072
`1.076
`0.4
`1.068
`1.069
`0.1
`
`0.998
`1.004
`0.6
`0.997
`1.004
`0.7
`1.003
`1.008
`0.5
`1.003
`1.011
`0.8
`0.851
`0.857
`0.7
`0.860
`0.866
`0.7
`1.169
`1.177
`0.7
`1.158
`1.161
`0.3
`
`1.001
`1.012
`1.1
`1.001
`1.013
`1.2
`1.011
`1.019
`0.8
`1.012
`1.025
`1.3
`0.759
`0.754
`0.6
`0.775
`0.783
`1.0
`1.316
`1.324
`0.6
`1.294
`1.295
`0.1
`
`1.007
`1.024
`1.7
`1.008
`1.032
`2.3
`1.029
`1.041
`1.1
`1.035
`1.055
`1.9
`0.625
`0.632
`1.1
`0.648
`0.656
`1.2
`1.618
`1.624
`0.4
`1.571
`1.568
`0.2
`
` Field size
`(x1,x2,y1,y2)
`5,5,5,5
`
`Depth
`(cm)
`1.5
`
`10.0
`
`1.5
`
`10
`
`1.5
`
`% difference
`
`% difference
`10,10,10,10
`
`% difference
`
`% difference
`5,5,0,10
`
`% difference
`
`% difference
`5,5,0,10
`
`% difference
`
`% difference
`
`10.0
`
`1 (calc.)
`
`1.5
`
`10
`
`2 (calc.)
`2(meas.)
`
`2 (calc.)
`2 (meas.)
`
`2. Physical Wedge Emulation
`Physical wedge emulation is the most common method used in non-physical wedge TPS
`implementation. The non-physical wedges are made to emulate the corresponding physical
`wedges in the TPS. Because of some intrinsic differences between the physical and non-physical
`wedges the latter cannot emulate the former in all aspects. Users must take extreme precautions
`in this emulation approach to ensure the safe and accurate clinical application. The differences
`between physical and non-physical wedges include the unique wedge factor variation with field
`size and wedge angle, and the lack of depth dependence of wedge factor due to the absence of
`beam hardening effect in non-physical wedged beams. A number of commercial TPS have
`incorporated EDW using this emulation method: they include, CMS, ROCS, Pinnacle, and
`TheraPlan. Depending on the specific requirements of each TPS, the beam data required for non-
`physical wedge TPS implementation could be different. They include a) non-physical wedge
`beam profiles of different wedge angles, field sizes, and depths; b) wedge factor of different
`wedge angles, field sizes, and off-axis distance, or c) wedge filter files which specify the physical
`description of the wedge filters.
`The wedge filter files can be generated with the following method based on measured
`non-physical wedge data. Measure beam profiles W (x, y) along wedge direction x at a fixed
`depth and SSD and at different off-axis distance y (the central axis is at x = y = 0). Only one
`direction of off-axis is needed because of the wedge symmetry. Measure the corresponding open
`field profiles O(x,y) and calculate the ratio of the wedged and open profiles,
`
`Page 6 of 13
`
`

`
`Page 7 of 13
`
`Chang & Gibbons 1999 AAPM RC
`
`R(x,y)|SAD,d = W(x,y)|SAD|/ O(x,y)|SAD,d
`
`where SAD = SSD +d
`
`If the filter to source distance on the accelerator is FSD, the virtual wedge filter file f (x', y', d')|SFD
`can be calculated based on the measured data above and a given linear attenuation coefficient µ
`value.
`
`x' = (SFD/SAD) x;
`
`y' = (SFD/SAD) y;
`
`The thickness of the filter at location (x', y'), d', can be calculated using a simple exponential
`function:
`
`R(x,y) = e - µ d'/cos θ
`
`where tan θ = [x'2 + y'2] 1/2 / SFD
`
`The cosine term basically solves the beam divergence issue in the filter file calculation.
`
`In contrast to the physical wedges, both the VW and the EDW have different asymmetric field
`size limitations for different wedge directions due to the constraints of the jaw travel range as
`described previously (Table 1). For example, for an asymmetric 20x30 (30 = 10, 20) field only
`one wedge direction is possible in the long dimension of the field. The opposite wedge direction
`is not possible because the moving jaw can only travel across the central ray 10cm. It is very
`desirable to install this field size limitation into each wedge filter file if possible. The wedge filter
`files can also be created from the STT table with good results [Klein 1997].
`Although physical wedge emulation methods enable almost any "closed" commercial TPS to
`adopt non-physical wedges, the challenge lies in the wedge factor computation, which is quite
`different than that of the physical wedges. In addition, one needs to fully understand those
`aspects of the non-physical wedge that cannot be emulated in the TPS. For example, non-
`physical wedges do not have the beam hardening effect physical wedges often possess therefore
`no such correction is needed.
`
`3. Synthesis of two or more wedge fields
`This type of non-physical wedge implementation is not very different from the physical
`wedge emulation methods. It uses beam data from an open field and one or more wedged fields
`to synthesize a non-physical wedge of any angle up to the largest wedge angle which beam data is
`used for the synthesis. Compare to the method of physical wedge emulation this method requires
`fewer beam profile measurements for TPS commissioning. EDW is ideal for such an
`implementation method considering it is intrinsically composed of an open field and a 60-degree
`wedged field with appropriate weighting. Several wedge angles are reportedly used to synthesize
`VW beam profiles. The concern with this multiple wedged beams method is that it may introduce
`complexity in wedge factor calculation since wedge factor calculation must be synthesized as
`well. In fact, it may not be a far reaching idea to bypass the complications encountered in the
`simulation of the non-physical wedges in the TPS altogether by using a combination of open and
`a 60-degree wedged fields of appropriate beam weighting. In this manner, the advantage of the
`treatment automation is preserved as well as the simplicity and accuracy of the conventional
`treatment planning technique.
`
`Page 7 of 13
`
`

`
`Page 8 of 13
`
`Chang & Gibbons 1999 AAPM RC
`
`IV. Non-physical wedge TPS commissioning and measurement tools
`
`The method of the TPS commissioning depends on the way the non-physical wedge is
`implemented into the TPS. Once the implementation method is chosen the commissioning should
`provide all necessary beam data and the verification of the TPS output accuracy in terms of all
`clinically relevant parameters. A variety of different treatment situations should be evaluated
`during the commissioning to verify the accuracy of the TPS, and to identify the circumstances
`when the implementation method fails to give correct answers. Such verification is especially
`crucial for the physical wedge emulation and synthesis implementation methods, which cannot
`correctly simulate the non-physical wedge functions in all aspects.
`Prior to the TPS commissioning the non-physical wedge function itself must be commissioned
`first. This commissioning includes the verification of the accelerator output rate variation as a
`function of jaw motion, wedge angle, field size, and other relevant parameters, regardless of if the
`function is governed by an analytical equation (VW) or a STT (EDW).
`The photon fluence modeling type of implementation method requires the least amount of
`data collection for both the VW and EDW. Standard beam data collection, which is normally
`used for static open field treatments, are used for the photon fluence modeling type of TPS
`implementation. Almost no special beam data (non-physical wedge beam) collection is needed for
`data input to the TPS. However, non-physical wedge data collection is still indispensable for TPS
`implementation verification.
`Non-physical wedge beam profile and wedge factor measurements are often required by the
`physical wedge emulation and synthesis methods. Beam profile measurements of a dynamic
`treatment can be done rather conveniently using commercially available multi-detectors array
`systems but it is also doable using standard dosimetry equipment. The Profiler™ diode-array
`measurement system is an ideal tool for dynamic treatment measurements in commissioning and
`routine QA. Besides collecting the conventional cumulative dose information the Profiler system
`is also capable of collecting time-dependent information, which can be used to measure both the
`collimator jaw speed (figure 8) and the output rate variation during VW irradiation. An ion
`chamber array detector system by Wellhöfer is also commonly used for the commissioning
`measurement. A simple technique of using a single ion chamber [Bayouth & Steinberg] can
`produce the above results with good accuracy.
`
`Figure 8. Virtual wedge jaw speed measured
`by Profiler™ detector array using a special
`time-dependent measurement mode. Beam
`profile was collected every second during
`VW irradiation. The moving field edge,
`represented by the point on the beam
`penumbra where the slope is the largest, was
`analyzed as a function of time. The
`measured jaw speed was within 1% of the
`expected value of 4.0 mm/sec in this case.
`
`Page 8 of 13
`
`

`
`Page 9 of 13
`
`Chang & Gibbons 1999 AAPM RC
`
`V.
`
`Small field size and photon source model effects
`
`Non-physical wedge treatments involve irradiation through very small and ofien off-axis
`openings that are not frequently encountered and therefore may not be well-evaluated in
`conventional static treatments. Dynamic wedge treatment uses field widths as small as 5 mm and
`10 mm for virtual wedge treatment. The accuracy of the dose and MU calculation for these small
`field width situations is highly dependent on the accuracy of head scatter data in this region and
`on the photon source model used in the TPS. Head scatter data must be directly measured in the
`small field region instead of extrapolating data of larger field sizes (even 4cmx4cm). Significant
`error can occur in MU calculation of the small segment fields if one extrapolates head scatter data
`from fields where lateral electronic equilibrium condition is satisfied to small fields where the
`condition does not exist. Figure 9 shows the dose decrease with reduced field width due to both
`the lack of lateral electronic equilibrium in the measurement media and the reduced photon
`fluence from the source. Figure 9 clearly shows that a simple linear extrapolation from data of
`field size above 4x4 can severely over-estimate the dose, or under-estimate the MU required to
`deliver a given dose in the narrow field situations. The reduction of photon fluence in narrow and
`off-axis field situations is illustrated by Figure 10. Only a portion (shaded area) of the photon
`source Volume (indicated by a sphere) is "seen" from the measurement location under the narrow
`and off-axis field. The amount of the source "seen" or the amount of photon fluence at the
`measurement location depends on the jaw settings and the measurement location itself.
`Obviously the reduction of the photon fluence in Figure 10 is highly source model dependent.
`The TPS photon source model which describes the intensity distribution of the photon source in
`the accelerator head should be modified and verified so there is a good agreement between the
`calculated and measured dose in all field configurations. The issue of small field width and
`photon source model primarily affects the "toe" end (the high dose end) of the wedge-like beam
`profile only, which is where the narrow field irradiation contributes in non-physical wedge
`treatments.
`
`Figure 9
`
`Figure 10
`
`Siemens HXE 6r1V
`
`1.2
`
`0.8
`
`0.6
`
`0.4
`
`collimator
`
`jaws
`
`as
`Qj
`
`XGTi
`
`n!
`
`field width (length = locm)
`
`0U
`
`,QQ'
`
`\u-
`U1
`LLH-I
`CI
`93
`
`OG
`
`0.2
`
`0
`
`I
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`Meas. point
`
`Page 9 of 13
`
`

`
`Page 10 of 13
`
`Chang & Gibbons 1999 AAPM RC
`
`VI. Advantages of non-physical wedges
`
`Treatment delivery automation is the most apparent and significant advantage of the non-
`physical wedge functions. Other advantages over the physical wedges include increased field size
`(40 cm) in the non-gradient direction and reduced peripheral dose. The latter can bring about a
`clinically significant outcome for tangential breast irradiation of young women, who can develop
`a radiation-induced malignancy in the contralateral breast due to the peripheral dose. Figure 11
`shows that virtual wedge treatments produce the least contralateral breast dose compared to the
`physical wedge and other treatment techniques in a humanoid breast phantom study by Chang et
`al. [1999]. Li & Klein [1997] showed that DW upper jaw wedging produce the same peripheral
`dose as the open field.
`The treatment delivery automation also allows the user to achieve simple forms of intensity
`modulation within the treatment port for dose distribution improvement and even for dose
`delivery error reduction in matching fields treatment [Shackfors & Bjarngard 1996].
`
`Figure 11. Contralateral breast dose in tangential breast irradiation. TLD chips were used
`at different locations in the contralateral breast. The vertical axis displays the ratio of the
`measured contralateral breast dose to the treatment dose [Chang et al. 1999].
`
`VII.
`
`Issues and concerns of non-physical wedge in clinical application
`
`A smooth clinical application of non-physical wedge modalities in an often busy and
`complex radiation therapy environment requires reliability, flexibility, and simplicity. The great
`advantages of a new technology brought about in one aspect of the operation is always
`compensated by some drawback it inevitably introduced. Non-physical wedges are certainly no
`exception. Some the drawbacks are listed below.
`
`Page 10 of 13
`
`

`
`Page 11 of 13
`
`Chang & Gibbons 1999 AAPM RC
`
`a) For a multi-accelerator department having only one accelerator equipped with the non-
`physical wedge, the non-uniformity in accelerator capabilities can create confusion and
`difficulties when swapping patients from one accelerator to the other. The large difference in
`wedge factor value between the non-physical and physical wedge can result in significant
`under/overdoses by simple mistakes. VW has certain advantages over EDW in this regard
`because of its near unity wedge factors.
`The accelerator heterogeneity problem is especially a concern if any of the non-physical
`wedge features which can not emulated by physical wedges are used. They include arbitrary
`wedge angles or wedge angles other than those available in a physical wedge, large field sizes
`in non-gradient direction and large asymmetric fields.
`
`b) Non-physical wedges have complex correlation among field configuration, wedge angle, and
`the MU required for each accelerator. The correlation depends on the limits on the output rate
`variation and jaw speed variation. These limits on field size and wedge angle configuration
`are difficult for a TPS to predict therefore to avoid planning the treatments which cannot be
`delivered at the accelerator. Siemens has developed an Excel program (runs on both Macs
`and PCs) called 'Virtual Wedge Simulation Spreadsheet" [Siochi]. The spreadsheet simulates
`the actual behavior of the accelerator and thus predicts if a given input treatment is deliverable
`before patient treatment.
`
`c) Many accelerators equipped with the non-physical wedge function also have MLC. The two
`automatic functions together significantly increase the level of treatment delivery automation.
`However, the wedge direction is predetermined once the orientation of MLC collimator is
`chosen for optimal treatment port definition, and vice versa. Milliken et al [1998] reported
`that 25% of the head & neck and lung cases studied required an average of 20-degree
`difference between the wedge and the MLC directions to achieve the optimal result.
`
`d) Online portal imaging is preferred at the beginning of irradiation for treatment setup
`verification. EDW does not interfere with online imaging since the treatment field opening
`changes from large to small during treatment delivery. However, a VW treatment, whereby
`the field opening sequence is from small to large, interferes with the online portal imaging.
`McGhee et al [1997] offered a solution to this problem by using a combination of open and 60
`degree VW angle fields.
`
`Non-physical wedge modalities have the capacity to offer something more than merely
`elimination of the manual handling of the physical wedge during treatment. The treatment
`delivery automation of the non-physical wedge together with other automation features of the new
`accelerators and of the new treatment record & verify system can greatly decrease the treatment
`delivery time per field. This reduction makes the many-fields treatments, designed by 3D
`conformal treatment planning for better clinical outcome, clinically feasible. The flexibility of
`the non-physical wedges should be used to improve the treatment dose distribution, such as using
`multiple wedged fields in the treatment port and multiple wedge orientations in the same
`treatment port. The latter can be especially helpful when MLC is used to define the treatment
`field.
`
`The authors can not guarantee the accuracy of the information, especially regarding the non-physical wedge
`functionality of the TPS. We apologize if we have made errors or omissions in citation.
`
`Page 11 of 13
`
`

`
`Page 12 of 13
`
`REFERENCES
`
`Chang & Gibbons 1999 AAPM RC
`
`Buyout John E. and Steinberg Todd H. Virtual Wedge Implementation on A Treatment Planning
`System (unpublished).
`Bank M., Desrosiers C., and Papiez L. Commissioning a Siemens Virtual Wedge [VW]
`(unpublished)
`Bank Morris, 1999 Implementation of a Siemens Virtual Wedge on RenderPlan Planning System
`AAPM presentation.
`Beavis AW, Weston SJ and Whitton VJ 1996 Implementation of the Varian EDW into a
`commercial RTP system Phys. Med. Biol. 41 1691-1704.
`Bengtsson M, Furre T, Rodal J, Skretting A and Olsen DR 1996 Measurement of Dynamic
`wedge angles and beam profiles by means of MRI ferrous sulfate gel dosimetry. Phys.
`Med. Biol. 41(2):269-77.
`Bidmead AM, Garton AJ and Childs PJ 1995 Beam data measurements for dynamic wedges on
`Varian 600C (6-MV) and 2100C (6- and 10-MV) linear accelerators Phys. Med. Biol. 40
`393-411.
`Chang S, Deschesne K, Cullip T, Parker S, Earnhart J 1999 A comparison of different intensity
`modulation treatment techniques for tangential breast irradiation Int. J. Radiat Oncol Biol
`Phys (submitted).
`Desobry GE, Waldron TJ and Das IJ 1998 Validation of a new virtual wedge model Med.
`Phys. 25 71-2.
`Das IJ, Steinberg TH 1998 Virtual wedge factor: an uncertainty analysis, Med. Phys 25(7):A204.
`Earley L 1997 Larger field sizes: an advantage of the dynamic wedge. Med. Dosim. 22(3):193-5.
`Gibbs G and Leavitt DD 1997 Commissioning the Varian Enhanced Dynamic Wedge using
`RAHD treatment planning system Med. Dosim. 22(3):227-9.
`Gibbons JP 1998 Calculation of enhanced dynamic wedge factors for symmetric and
`asymmetric photon fields Med. Phys. 25 1411-1418.
`Gibbons J and Vassy D 1998 Calculation of virtual wedge factors for symmetric and
`asymmetric photon fields Med. Phys. 25 A188.
`Huntzinger CJ 1993 Dynamic wedge: a physicist’s perspective Proceedings of the Varian
`Dynamic Wedge Users’ Meeting, Calgary, 1992 (Varian, Palo Alto, CA)
`Karlsson M 1997 Implementation of Varian's EDW 5.2 on a Clinic 2300C/D for use with Helax
`TMS 3.1 dose planning system. Med. Dosim. 22(3)215-8.
`Kijewski PK, Chin LM and Bjarngard BE 1978 Wedge-shaped dose distributions by computer-
`controlled collimator motion Med. Phys 5 426-9
`Klein EE, Low DA, Meigooni AS and Purdy JA 1995 Dosimetry and clinical implementation
`of dynamic wedge Int. J. Radiat. Oncol. Biol. Phys. 31 583-92.
`Klein E, Gerber R, Zhu R, Oehmke F and Purdy J 1998 Multiple machine implementation of
`enhanced dynamic wedge Int. J. Radiat. Oncol. Biol. Phys. 40 977-985.
`Klein EE 1999 MU calculations for dynamic and virtual wedges in Monitor Unit Calculations
`for External Photon and Electron Beams: Proceedings of the Southeast AAPM Chapter
`(Atlanta, 1999), (In Press)
`Klein EE 1997 treatment planning for enhanced dynamic wedge with the CMS focus/Modulex
`treatment planning system Med. Dosim. 22(3):312-4
`Leavitt DD, Martin M, Moeller JH and Lee WL 1990 Dynamic wedge field techniques through
`computer-controlled collimator motion Med. Phys. 17 87-91.
`Leavitt DD, Lee WL, Gaffney DK, Moeller JH and O”Rear JH 1997 Dosimetric parameters of
`enhanced dynamic wedge for treatment planning and verification Med. Dosim. 20 177-
`183.
`
`Page 12 of 13
`
`

`
`Page 13 of 13
`
`Chang & Gibbons 1999 AAPM RC
`
`Li Z and Klein EE 1997 Surface and peripheral doses of dynamic and physical wedges Int. J.
`Radiat Oncol Biol Phys 37(4):921-5.
`Liu C, Zhu TC and Palta JR 1996 Characterizing output for dynamic wedges Med. Phys. 23
`1213-1218.
`Liu C, Li Z and Palta JR 1998 Characterizing output for the Varian enhanced dynamic wedge
`field Med. Phys. 25 64-67.
`Liu C, Waugh B, Li Z, Zhu TC and Palta JR 1997 Commissioning of enhanced dynamic wedge
`on a ROCS RTP system. Med. Dosim. 22(3)231-6.
`Lydon JM and Rykers KL 1996 Beam profiles in the nonwedged direction for dynamic wedges
`Phys. Med. Biol. 41(7):1217-25.
`McGhee P, Chu T, Leszcezynski K and Dunscombe P 1997 The Siemens Virtual Wedge Med.
`Dosim. 22 39-41.
`Milliken BD, Turian JV, Hamilton RJ, Rubin SJ, Kuchnir FT, YU CX, and Wong JW 1998
`Verification of the omni wedge technique. Med. Phys.25(8):1419-23.
`Papatheodorou S, Zefkili S, and Rosenwald JC 1999 The 'equivalent wedge' implementation of
`the Varian Enhanced Dynamic Wedge (EDW) into a treatment planning system Phys.
`Med. Biol 44(2):509-24.
`Salk J, Blank P, Machold U, Rau E, Scheider E, Röttinger E. 1999 Physical Aspects in the
`Clinical Implementation of the Enhanced Dynamic Wedge (EDW) (unpublished)
`Samuelsson A, Johansson KA, Mattsson O, Palm A, Puurunen H, Sernbo G 1997 Practical
`implementation of enhanced dynamic wedge in the CadPlan treatment planning system.
`Med. Dosim. 22(3):207-11.
`Santvoort JV 1998 Dosimetric evaluation of the Siemens Virtual Wedge Phys. Med. Biol. 43(9),
`2651-2663.
`Shackford H and Bjsrngard BE 1996 A dynamic match-line wedge Int. J. Radiat Oncol Biol
`Phys 35(1):161-3.
`Siochi 1998, Virtual wedge spreadsheet, Internal document of Siemens Medical Systems, OCS.
`Thomas SJ and Foster KR 1995 Radiotherapy treatment planning with dynamic wedges—an
`algorithm for generating wedge factors and beam data Phys. Med. Biol. 40 1421-1433.
`Van Santvoort J 1998 Dosimetric Evaluation of the Siemens Virtual Wedge Phys. Med. Biol.
`43 2651-2663.
`Waldron TJ, Boyer AL, Wells NH and Otte VA 1994 Calculation of dynamically-wedged
`isodose distributions from segmented treatment tables and open-field measurements
`Med. Phys. 21 873.
`Weber L, Ahnesjo A, Nilsson P, Saxner M, and Knoos T 1996 Verification and implementation
`of dynamic wedge calculations in a treatment planning system based on a dose-to-energy-
`fluence formalism. Med. Phys. 23(3):307-16.
`Weides CD, Mok EC, Chang WC, Findley DO and Shostak CA 1995 Evaluating the dose to the
`contralateral breast when using a dynamic wedge versus a regular wedge Med. Dosim.
`20(4):287-93.
`Zhu TC, Ding L, Liu CR, Palta JR, Simon WE, Shi J 1997 Performance evaluation of a diode
`array for enhanced dynamic wedge dosimetry Med. Phys. 24(7): 1173-80
`
`Page 13 of 13