MedicalDosmetry, Vol. 13. pp. 25-27 Printed in the U.S.A. All rights reserved.
t Ml
0739-021 l/88 $3.00 Copyright 0 1988 American Association of Medical Dosimetrists
POLYETHYLENE-LEAD TISSUE COMPENSATORS MEGAVOLTAGE RADIOTHERAPY*
FOR
JAMESSPICKA,M.S.,’ KERRY FLEURY, B.S.,2 and WILLIAM POWERS, M.D.2 ‘Now at St. Joseph Hospital, 302 Kensington, Flint, MI 48502 2Harper-Grace Hospitals, Gershenson Radiation Oncology Center, 3990 John R Street, Detroit, MI 4820 1 Abstract-Tissue compensators afford one the opportunity of producing idealized dose distributions in radiotherapy. We have developed a technique which may be used to produce fast, accurate and inexpensive compensators within a few minutes and requires minimal patient involvement. Three dimensional contour data is acquired using a moire fringe photograph taken at the time of simulation. The photograph is projected to the size needed for the actual compensator and a sketch of the compensator is made. The sketch is placed in a specially designed small portable pantographic unit and the design is traced while the unit cuts the compensator from an indexed polyethylene-lead blank which is premounted on an acrylic tray. The polyethylene-lead material proves to have an ideal combination of properties for compensator construction including relatively high density, ease of machining, good handling characteristics and low cost. Key Words:
Compensator,Megavoltage,Moire
To design a compensator the photo is optically projected to match the actual dimensions of the compensator and the ISO-SSD lines are traced. Thus the compensator pattern evolves directly from the beams eye moire photograph and the ISO-SSD lines become indicators of compensator steps.
INTRODUCTION
Irregularities in three dimensional patient contours introduce the largest single source of error in current radiotherapy treatments. Dose variations of up to 25% are common at midplane in a patient due to variations in patient thickness across a treatment field. Computerized dose distributions, while helpful, provide accurate information only in the plane of calculation and actual doses a few centimeters away may be radically different. Tissue compensators afford one the opportunity of providing idealized dose distributions for three dimensional radiotherapy. We have developed a technique which allows one to produce accurate and inexpensive compensators in a few minutes and which requires minimal patient involvement.
COMPENSATOR
DESIGN
The first step in compensator design is determining the thickness of compensator material which corresponds to the nominal 1 cm steps indicated by the moire photograph. To provide a beam of uniform intensity at a selected reference plane in the patient one can sequentially use the following equation to calculate the thickness of material which corresponds to each step indicated by the moire photograph. X = (ln(TMR D/TMR(D - l)))/(-PC)
ACQUIRING
PATIENT
DATA
where,
Our system uses a compact moire fringe camera developed by A. Boyer and M. Goitein’ to acquire patient data at time of simulation. Patient involvement is thus limited to the taking of a beams eye photograph (Fig. 1) of the patient during simulation. Simulation time is increased by no more than five minutes. The beams eye photograph provides one with ISO-SSD lines superimposed on a view of the treatment portal. The photo then contains all the basic three dimensional information necessary for compensator construction.
* Presented
at the 1984 AAPM Meeting
in Chicago,
X = Compensator
step thickness
TMR D = TMR for reference depth D (cm) TMR(D - 1) = TMR for the adjacent compensator step indicated by the moire fringe photo 1 = The attenuation coefficient for polylead for the megavoltage beam of interest Routinely one can avoid any tedious manual calculation of compensator increments by using a precalculated table as shown in Table 1. Compen-
Ill. 25
Medical
26
Volume
Dosimetry
13, Number
1, 1988
Fig. 2. View of compensator cutter showing a finished compensator and compensator pattern. Fig. 1. Beams eye moire fringe photograph taken of a treatment portal.
The tray is then locked into a specially designed compensator cutter (Fig. 2). Into this device, resembling a pantographic router, is also placed a copy of the compensator pattern. By tracing the pattern one can cut the needed steps and plateaus into the polyethylene-lead material. The compensator cutter is designed to be used as a mechanically linked indexed system. This means that by placing the pattern in the correct location the compensator will be cut precisely into the poly-lead blank in the proper location as the pattern is traced. The compensator (Fig. 3) is ready to install in the treatment machine immediately upon removal from the cutter. Fabrication time averages 20 min.
sator increments exhibit energy, depth and field size dependence. Compensator design typically takes 20 min. FABRICATION
The compensators are constructed of polyethylene-lead. This easily machinable thermoplastic contains a homogeneous mixture of minute suspended lead particles. The quantity of lead is sufficient to increase the density of the material to 3 g/cm3. The material is cast in 1” thick slaps and provides the medium to compensate for up to 12 cm of oblique incidence. To fabricate a compensator, a block of material is mounted on a polycarbonate compensator tray.
VERIFICATION
The accuracy of the system was initially tested by constructing a convex tissue equivalent phantom that
Table 1. Table of compensator increments based on the ratio of the TMRs for 4 MV X-rays. Varian
CLINAC
4- 100,4
MV X-rays compensator
increments
Field size Depth
4x4
6X6
8X8
10x
IO
12 x 12
15 x 15
20 x 20
25 X 25
30 x 30
52 69 81 83 86 89 92 96 99 102 105 108
50 67 74 81 83 85 88 90 92 94 96 98
45 59 71 74 76 79 81 84 86 88 91 93
40 57 64 69 71 73 75 78 80 82 84 86
38 55 62 66 68 70 72 74 75 77 79 81
35 52 61 63 65 66 67 69 71 73 75 76
All thicknesses on the above chart are in thousandths Material attenuation coefficient = .168/cm. Comp. tray factor = .962.
of an inch.
3 4 5 6 7 8 9 10 II 12 13 14
71 101 110 115 116 117 118 119 120 121 122 123
62 86 98 103 105 107 109 111 112 115 117 120
57 76 86 93 95 98 101 103 105 108 110 113
Polyethylene-lead tissue compensators 0 J.
SPICKA et al.
Fig. 5. Scans showing: A. The dose profile at depth resulting from a normal phantom surface, B. The impact in the dose profile resulting from the introduction of the convex phantom surface and C. The dose profile at depth (with the convex phantom surface) after insertion of the compensator.
bution improves with compensation. Tests with other megavoltage modalities and phantoms have yielded equally positive results. COST The cost of compensators as follows:
Fig. 3. Finished compensator ready to install in the treatment unit.
may be broken down
Fixed Costs per Compensator could be placed on the surface of a conventional water phantom (Fig. 4). A moire photograph of the phantom was taken and a compensator was designed and constructed to provide an even dose at a reference depth in a beam of Co-60 radiation from an AECL Theratron 780. A series of three scans were then performed (Fig. 5). Scan A reveals the radiation intensity in the water phantom without the convex phantom in place. Scan B shows how the dose distribution is influenced by the addition of the convex phantom surface. Scan C reveals how the dose distri-
Material Film Cutters
Initial Set-Up Costs Moire Fringe Camera Compensator Cutter Reusable Trays
CLINICAL
Reference Probe
Ad;:t?;‘.jgula~
i_
Scan Depth Co 60 Field Size 20cm x 8cm at 80cm S50
Fig. 4. Experimental
set-up for compensator
$15.00 2.00 2.50 $19.50
evaluation.
1
$8,500.00 5,ooo.oo 500.00 $14,000.00 RESULTS
This system has been used to fabricate compensators used thus far in the treatment of approximately 1500 patients. Breast and lung cancer patients have predominated in the clinical applications and physicians have reported reduced side effects in both groups of patients. Clinical acceptance has been excellent. REFERENCE 1. Boyer, A.; Goitein, M. Simulator mounted moire tomography for constructing compensator filters. Medical Physics 7(I): 19-26; 1980.