ML20073E094

Treatment Room Study for Washington State Univ Boron Neutron Capture Therapy

ML20073E094
Person / Time
Site:	Washington State University
Issue date:	08/19/1994
From:	Ristow R AFFILIATION NOT ASSIGNED
To:
Shared Package
ML20073E091	List: ML20073E088 ML20073E094 ML20073E101
References
NUDOCS 9409280233
Download: ML20073E094 (22)

Text

.

1 J

1 i

l TREATMENT ROOM STUDY FOR WASHINGTON STATE UNIVERSITY i BORON NEUTRON CAPTURE THERAPY Rodney Ristow  ;

I August 19, 1994 l Student Research Participation Program l l

Associated Western Universities Incorporated l Funded by Idaho National Engineering Laboratory EG&G Idaho, Inc.

Idaho Falls, ID 83415 l

l l

l 9409280233 940914 PDR ADOCK 05000027 P PDR

h t

t ABSTRACT A shield wall study has been perfonned for shielding around an epithermal neutron beam I opening at the Washington State University Reactor Facility. This beam is not currently available, but '

a feasibility study has been performed which indicates that the existing thermal column could be. ,

modified and a beam comparable to the epithennal neutron beam at the Brookhaven Medical Research Reactor could be achieved. Using the output from the thermal column calculations, and i further calculations again using the two-dimensional DORT discrete-ordinates computer code, a  :

shield design has been developed which would both restrict access and minimize radiation dose to  ;

occupied areas around the thermal column. This shield wall would enclose a proposed treatment room into which the beam would be directed. In addition, consideration was given to minimizing the .

radiation levels inside the proposed treatment room for the benefit of the patient.

1 1

iii

1 1

I CONTENTS ABSTRACT........................,,,,,,,,,,,,,,,,,,,,,,,,,,,,ggg

1. INTRODUCTION............................................. 1 2.

ANALYSIS................................................. 2 {

2.1 Materials Considered............................... 2 2.2 Beam Insert........................................ 2 2.3 Use of Heavy Concrete.............................. 2 2.4 Room Lining........................................ 9 2.5 Use of Lead........................................ 9 2.6 The Ideal Mode 1.................................... 9 2 . 7 Co s t De te nmi na ti on . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 '

3.

CONCLUSIONS............................................. 15 4.

RE FE RE NCE S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 8 l

FIGURES 4

1. Reactor Model 34 BNCT.................................... 3

2. Filter Model E4.......................................... 4

3. Calculated Neutron Spectrum for BNCT Case E4............. 5

4. Beam Insert.............................................. 7

5. Beam Insert Installed.................................... 8

6. Ideal Room.............................................. 10

7. Plot of radiation into room (radial).................... 11 i

8. Plot of radiation out of room (radial).................. 12

9. Plot of radiation out of room ( axi al ) . . . . . . . . . . . . . . . . . . . 13 I

l V i i

1 l

I TABLES

1. Beam Characteristics.................................... 5

2. Material Costs.......................................... 5

3. Radia tion Levels for each Model . . . . . . . . . . . . . . . . . . . . . . . . . 6

4. Ideal Model Costs...................................... 14

5. Model Descriptions..................................... 14 l

6. Material Summary for each Model........................ 16

7. Cost Summary for each Model............................ 17 c

t i

y l

1 1

Vii l

l 1

1

l1 TREATMENT ROOM STUDY FOR ,

WASHINGTON STATE UNIVERSITY BORON NEUTRON CAPTURE THERAPY (BNCT) i

1. INTRODUCTION Several pharmaceutical drugs are currently available which seek out and attach themselves to rapidly reproducing cells, i.e.

cancer cells. Boron can be added to these drugs during fabrication and then the Boron could also be attached to the cancer cells. Once the Boron has been attached to the cancer cells they can be irradiated with neutrons which will then destroy the cancer cells while leaving the healthy tissue around the cancer unharmed. This is accomplished through neutron capture of the boron yielding a destructive, but short lived, alpha particle.

This procedure is currently being performed in the United States on animal patients and there are many interested parties who  ;

would like to see this type of treatment become available to human i patients. This treatment is presently being performed on human patients in Japan for a minimum cost of $60,000.00. Japan is the only country performing this procedure at this time and they have been the only one for about 20 years . 2 Problem: Design a workable plan to set up a cancer treatment facility at the Washington State University nuclear radiation ,

center. To safely provide a directed epithermal neutron beam that '

will be effective in destroying the cancer cells and at the same time not increase unduly the radiation risk to other personnel.

I. The filter design to provide the epithermal neutron beam is being worked on by Floyd Wheeler and David Nigg of INEL (EG&G).

This involves the use of an AlF3/Al material developed in Helsinki Finland. The achievable beam would be comparable to the epithermal neutron beam at the Brookhaven Medical Research Reactor (presently the best epithermal beam in the world)2, II. I have been working on shield designs for the treatment room itself. I have used as an input the output beam calculations from Floyd Wheeler's filter design. The computer calculations I have used employ the two-dimensional DORT code. .

Design Criteria: The shielding must reduce the neutron and gamma flux to below the maximum local levels and below .5 mrem /hr.

Cost of materials should be considered as well as availability of materials.

1.The Spokesman Review, Wed. May 4, 1994, " DOE bolsters cancer studies"

2. Reference 9 1

2. ANALYSIS The following information is based upon a shield wall study performed using the two dimensional DORT discrect-ordinates computer code. The input for these calculations was the output from earlier DORT calculations done for the filter design. These used reactor model 34 BNCT (see figure 1) and exterior beam filter design E4 with 65 cm AIF3/Al and Al2O3 outer perimeter (see figure 2). The calculated neutron spectmm for this combination is shown in figure 3. The comparison of the neutron beams for different facilities is given in table 1.

2.1 Materials Considered The materials considered for the shield design are commonly used for various shielding -

applications and are readily available through companies such as Reactor Experiments or local contractors. These included Borated Polyethylene, Lithiated Polyethylene, heavy concrete, and lead.

The costs for these materials, which were used to determine model costs, are given in table 2.

2.2 Beam Insert The greatest change in radiation levels outside the shield took place, not with varying the shielding materials, but with the variation of the beam inten,ity entering the room. A combination of smaller beam opening size and a simulated head reduced the inside room neutron radiation by a factor of over six (compare models 4 and 17 of table 3). By completely covering the beam opening with water, the inside room neutron radiation was, again, reduced by a factor of over five (compare models 23 and 24 oftable 3). The complete covering of the beam opening with water was done to simulate a beam insert erTect. At Brookhaven, beam inserts are currently available which fit the shape of patients' heads. This has the dual effect of reducing the body radiation to the patien+, himself, and reducing the radiation to those outside the shielding. Actual neutron dose inside the room dropped from 130.9 Rem / hour to 3.29 Rem / hour because of these modifications. Figures 4 and 5 show a theoretical beam insert by itself and inserted into the thermal column.

2.3 Use of Heavy Concrete Heavy concrete was chosen as the main shield material because ofits low cost and structural support.

It interacts with and reduces both the neutron flux and the gamma flux. Various thicknesses were tested. Findings proved that a back wall of two and a half feet and a side wall and ceiling of two feet would be adequate, in combination with the other shield materials, for reducing the neutron and gamma fluxes to less than one mrem / hour. The back wall needs to be thicker to account for the directional beam coming into the room.

2 1

1 2 3 4 5 6 7 8 9 f V V V V V V V V E* XXNXXXX E S

XX .

8 XX_X g

C*

S S F F F F S h

a D+ S F F F F F S h

E+ S S F F F F S s _ - . _

S 8 XX 8 8 e  :

a e XXXXXX :#

S Standard Fuel h " *** ' *'* F FLIP Fuel Reflector Block Irigure 1. Reactor Model 3411NC F 3

r HO 2 ordin -

2 y ,gg c, ;.j y,- .. .,3p _ . g -'.

m.__._. _

__- l <

j A13Oi '

Udnated

,# poly t [h ..,

. bismuth

> - $ p.

we -

y M Ij

' AF/M k, i M' J.~ ,_

'I wy. ,

-a l';. a #J

,l c.

!I vFKf6A -

" ~

L  :.

.$ f  :

v&?kkvMMMMM4 '

l P '- ,

.c, . .. t . re ?j u _

, b Y M

~

Thermal column assembly with A1F /A1 and A1 O filter (not to scale).

3 2 3 4

Figure 2.1 ilter Model E4 b

, , Ccs3 E4: 65-cm A!F/Al, exterior beam Al,0, at perimstsrs 1 p 4 l

3: 10' 3

lii

? -

~

10' f

E js 10

?:s ,

b j 10' I .

tot epi flux ::: 8.5x10' fast KERMA = 2.24E 11 10' 10" 10* 10' 10* 10* 10' 10' 10' 10' Energy (eV)

Figure 3. Calculated Neutron Spectmm for BNCT Case E4 Table 1. Beam Characteristics (from references 9 and 10)

Reactor Epithermal Flux Fast Neutron Gamma intensity KERMA Kf KERMA Kg (l.0E9 n/cm2-s) (1.0E-11 cGy/n-cm2) (1.0E-11 cGy/n-cm2)

BMRR 1.80 4.30 1,30 (3MW) i WSU -E4 0.846 2.24 1.20 (1MW)

HFR 0.33 10.4 8.40 (45MW) '

MITR-II 0.20 13.0 14.0 Table 2. Material Costs MATERIAL 1"x48"x48" 2"x48"x_ 4"x48"x36" B-Poly (201) $485.00 96" $2275.00 $1900.00 .

Li-Poly $1480.00 48" $3085.00 $4675.00 5 '

.I Table 3. Radiation Levels for each Model t t

1. total side side back back r oo.n room cost in wall wall wall wall neut. gamma

$ neutron gamma neutron gamma max max max out max ou' max out max out (z=550) (z=550)

(mr/h) (mr/h) (mr/h) (mr/h) (r/h) (r/h) 3 175980 0.670 1.704 1 9. 15 10.72 unavail unavail 4 167108 0.598 1.704 6.104 130.9 6.310 1.642 5 163913 0.660 1.702 6.106 6.309 130.9 1.641 6 138580 0 670 1.703 17,878 399.3 130.9 1.642 7 93463 0.537 1 2.060 5.482 6.704 131.3 2.438 8 99704 0.102 0.492 1.547 1.937 131.3 2.533 ,

10 26389 0.251 16.22 2.528 18.66 116.3 5.061 i 11 173349 0.113 0.623 1.492 1.852 107.3 1.780 12 173349 0.103 0.511 0.926 1.623 19.34 0.915 33 154839 0.103 0.512 1.921 2.036 19.34 0.915 14 118989 0.187 0.811 1.505 1.894 19.30 0.941 15 125159 0.187 0.812 1.505 1.894 19.31 0.944 1 16 118989 0.150 0.225 1.363 0.707 19.31 0.945 17 106483 0.392 0.168 4.619 0.487 19.28 0.910 18 112724 0.082 0.044 1.159 0.154 19.35 0.916 19 108187 0.397 0.171 1.164 0.155 19.31 0.913 20 114357 0.396 0.168 0.990 0.126 19.29 0.911  !

21 109298 0.400 0.168 0.307 0.053 19.27 0.929 22 95273 0.402 0.167 0.553 0.087 19.27 0.933 I 23 63108 0.137 0.051 0.766 0.110 19.36 0.977

{

24 63108 0.128 0.085 0.531 0.063 3.290 1.544 25 63108 0.128 0.005 0.645 0.372 3.290 1.544

{

6 l

l

. 1 l

i

$g?e1:yl Q \

> E m

Q) ,

a O  !

c- ,

y O

.g.

y -

b 3

'<t sh m/Ax - 0)

@ Z j

l C ,

\/

E , ;_ . '-  !

x  ;

O a>

CD y pr : @ o f

8 i s Y

g - P, o

u 1

1 t' I

,v

_b

>_ ,8

+

8 7

, =

,.4['-

..g' : * .;l.,_.reducedopening size ilA:

+.,

Front view of thermal column un beam insert Thencal Column Face

' ~- cam lawt

^

Side view of thermal column with beam insert Figure 5. Beam Insert Installed 8

l 2.4 Room Lining Within the room, several models were considered to determine if a lining would be used for the concrete walls. There are two benefits to a lining: first, the gamma dose inside the room reduces by a factor of three if one is used (comp:.re models 4 and 10 of table 3), and second, the concrete, itself, does not become activated from neutron interactions. Also, studies showed that the gamma dose produced by boron interaction in the Borated Poly raised the inside room gamma dose by a factor of about 1.5 over an identical model using Lithiated Poly (compare models 4 and 7 of table 3).

With respect to a lining, the study showed that the thickness of the lining did not seem to greatly I effect the radiation dose in the room. As long as some lining was there, the room dose went down.

As a result of these findings, a one inch Lithiated Polyethylene model was selected as the most cost-l effective and patient-beneficial model. I 2.5 Use of Lead Lead was considered for bordering outside the concrete to further reduce the gamma dose.

Lead bricks are fairly easy to obtain and could possibly even be donated by various facilities.

Additional structural support would have to be considered for the lead. However, the sharp decreases in gamma doses could well be worth the investment. The gamma dose for similar models, one with two inches oflead and one without any lead, showed a difference in radiation levels by a factor of 15 (compare models 14 and 23 of table 3). Lead sheets are available which are one halfinch thick. When compared, there was a radiation factor difference of six between the two lead thicknesses with equal thicknesses of concrete (compare models 24 and 25 of table 3).

1 2.6 The Ideal Model A shield model with a one inch Lithiated Polyethylene lining on the inside walls, two feet heavy concrete side walls, two and a half feet heavy concrete back wall, and two inch lead bdcks around the exterior of the room would provide the optimum shield design (see figure 6). This would minimize patient radiation, minimize external radiation from the room, and minimize material cost in constmetion. Figures 7,8, and 9 show the radiation plots for a model of these specifications. For a twelve by twelve foot room with this shielding the estimated meterial cost would be $63,000.00, assuming the lead bdcks would be donated. The cost breakdown for this is given in table 4. Note that this does not include the cost for a containment type door, nor were radiation effects for a door included in this study. For a detailed description of the other models used see table 5. Using this table and the radiation levels given in table 3 one can compare how the radiation changed for each change in specific models.

9

a t

e .

e r &

_ c

_ n o

C

_ y y t l

v r n d o a d e es r i

v o P U

e H

a e

t a mo no l w i ad n

t ope y l p)l.

- ct

$le]f1,* .

$ i d.%"

i g

i!e

_ y;tCwI N jg _

4!

)

y$j&m lll e

i l

_ }g:

'  : hE

,.2h  : . ,

1i1 i

a m c -

3 s s 4 -

o 3 i

t  : 8 t 2 -

o 3 n -

( &

n 3-s We

)

n e m 8 e t

e o

e n -

o m f s 2 n -

R a  :

n e 1 x h )t e

l B 2 c e a 1 n f e

+

r 4 5 2 d i o 1 5

a 2 w I r w e m -

t n s -

i

( n s

n 3

3 s

s . -

a -

- m 3

+

~ ;,.2b . , -x: -.

. , . w r m

o o

gaq)M vc-- .

l R +

a e

d yt I .

- 6.

c r

ug i

F -

-O -

.t' 4 ) .!

radial distribution of dose rates at Z=372 (Rin) 4 case 24: 10.5 cm Head,2in Pb,1in LiPoly,2.5/2ft concrete on back/ side walls 10 , , , ,

, -s .

t

\

l I

i -- ,t

-\'

\

2 10 -

.. s -

c \, ',

a '

s O \ '

.'s

~

.c ---- neutron u ~. s e gamma

a. .3

~

E N',

y 10 -

i

' ' 's ..'..',

s -

m s.

g

, ... , . . ~ .,

g '

2d,_. _ _ _ j O s a

Q '. s ,Ns s g

a 10'" -

'6Ns _ 2 C

~x n

%, -'5 ss s s 4

.s g N s ,

Q 10

-4 ' ' ' '

's s-3 I

0 50 100 150 200 250 s

- u

. R distance (cm) g u:

=

-w y

.,, w

radial distribution of dose rates at Z=800 (Rout)

Case 24: 10.5 cm Head,2in Pb,1in LiPoly,2.5/2ft concrete on back/ side walls 10-2 . , , .

, , . . . i neutron gamma E

s g-gos -

e a

Qm ----

& N -

s N

E \ s

~

m s

[

_ 's h10" -

N -

-f, e s 8 ' e o L- Ns 5 e

's -  %

~

s s

's S s

s a 10* ' ' ' ' ' ' ' ' ' ' \' ' ' '

0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 ,5 Distance from beam centerline (cm) [

k

Axial distribution of dose rates for outer surface of treatment room wall (Zoutb)

-2 case 24: 10.5 cm Head,2in Pb,1in LiPoly,2.5/2ft concrete on back/ side walls 10 . , , .

f e

8 10

.c m

---

neutron m.

o. gamma E

e T

rn m -

E u

8 4 's o 10 -

Ns -~~ -

g a js , ~ _ -

's g l\ ,'

s s '.

.i- -

s

,- t g

\ l (

a,- . . . .'

\ $

l. 4 i -s' i 3

\

-5 \ t a

t .' o ii e/

\

\\ t g

1l \

5. / , '

• i .=3

, .~

, g 10_, ' ' ' ' ' ' ' a 300 400 500 600 700 800 _$

- Z distance (cm) 9 es 3'

E z

Table 4. Ideal Model Costs Number of one inch Cost per total cost LiPoly sheets sheet 27 $1480.00 $39,960.00 Cubic feet of Cost per total cost heavy concrete cubic yard 1250 $500.00 $23,148.00 Lead Bricks - no cost calculated Total material cost $63.108.00 Table 5. Model Descriptions Model Beam back side back side and Lead size coner coner Poly Poly poly (cm) notes (cm) (cm) (cm) (cm) (cm) 3 Li 15 44 44 6 6 4 Li 15 44 44 15 6 5 Li 15 44 44 15/6 6 6 Li II 44 44 15 to 151cm 6 6 7 B 15 44 44 15 hole to 16cm 6

8 B '

.5 60 59 14 6 9 15 74 65 none none not used 10 15 74 65 none none head hereafter 11 Li 15 60 59 14 6 12 Li 8.25 60 59 14 6 13 Li 8.25 60 59 8 6 14 Li 8.25 60 59 10/4 3 10 to 106cm 15 Li 8.25 60 59 10/4 3 16 Li 8.25 60 59 10/4 3 2 Li on front 17 Li 8.25 45 45 10/5 5 18 Li 8.25 60 60 5 10 to 67cm 10/5 5 5 19 Li 8.25 60 45 10/5 5 5 20 L1 8.25 60 45 15/5 5 21 Li 8.25 75 45 5 15 to 67cm 22 Li 10/5 5 5 10 to 67cm 8.25 75 45 5 5 5 23 Li 8.25 75 60 2.5 2.5 5 24 Li 8.25 75 60 2.5 25 Li 8.25 75 60 2.5 2.5 5 hole fully covered 2.5 1.25/5 1.25 on back wall i

14 '

• l 1

I i

2.7 Cost Determination l To determine the material costs for the various models the following assumptions were made:

1. The Poly sheets could be cut in length or height, but not in width, thus 1",2", or 4" thick sheets would have to be used.

2. For a given model, even if the dimension is just over the inch thickness, the next thicker size would have to be used for the calculation to be valid.

3. Concrete could be poured in any thickness. A proportional cost to that per cubic yard was used.

4. The walls would be eight feet high.

5. The inner dimension of the room would be 12' by 12'.

6. 5 cm is about 2 inches,10 cm is about 4 inches, etc.

7. The ceiling thicknesses and materials would be the same as the side walls Us~mg these assumptions tables 6 and 7 were generated, which show for each model the exact number of polyethylene sheets and feet of concrete used. Table 7 gives the cost summary for each model.

3. CONCLUSIONS The ideal model, as shown in figure 6, and with a cost of $63,108.00, would provide the optimum in minimizing cost and radiation levels. A beam insert should be used in conjunction with this design to yield the rne calculated radiation levels. A further study, which would include

)

radiation effects for a door, must still be performed. The costs for such a door will increase the '

material costs of the room as well. The actual constmetion costs will vary and depend on local  ;

contractmg. '

l l

15 l

1

l O

l l

Table 6. Material Summary for each Model '

1. 4" sheets 2" sheets 1" sheets feet of concrete sides /back/ top  !

sides /back/ top sides /back!!op sides /back/ top '

3 12!6/9 12/6/9 300/192/336 4 0/8/0 12/6/9 12/0/9 300/192/336 5 0/7/0 12/6/9 12/1/9 300/192/336 6 12/6/9 12/6/9 300/192/336 7 0/8/0 6/3/5 12/0/9 300/192/336 8 0/8/0 6/3/5 12/0/9 400/272/493 10 466/352/607 11 0/8/0 12/6/9 12/0/9 400/272/493 l 12 0/8/0 12/6/9 12/0/9 400/272/493 13 0/8/0 12/0/9 12/0/9 400/272/493 14 0/5/0 12/3/9 400/272/493 15 0/5/0 12/3/9 2-front 400/272/493 16 0/5/0 12/3/9 400/272/493 17 0/3/0 12/4/9 300/192/336 18 Or3/0 12/4/9 400/272/493 19 0/3/0 12/4/9 300/272/348 20 0/3/0 12/6/9 300/272/34B 21 0/3/0 12/4/9 300/320/360 22 12/6/9 1

300/320/360 23 12/6/9 400/340/510 24 12/6/9 400/340/510 25 12/6/9 400/340/510 l

16

d' Table 7 Cost Summary for each Model

1. # of 4" # of 2" # of 1" feet of sheets / cost total '

sheets / cost sheets / cost concrete / cost cost in S

1 3 B/$37400 27/$83295 27/$39960 828/$15333 175980 j

4 8/$37400 27/$83295 21/$31080 828/$15333 167100 5 7/$32725 27/$83295 22/$32560 828/$15333 163913 6 27/$83295 27/$39960 828/$15333 138580 i 7 8/$15200 14/$31850 21/$31080 828/$15333 93463 8 8/$15200 14/$31850 21/$31080 1165/$21574 99704 10 1425/$26389 26389 11 8/$37400 27/$83295 21/$31080 1165/$21574 173349 I 12 8/$37400 27/$83295 21/$31080 1165/$21574 173349 13 B/$37400 21/$64785 21/$31080 1165/$21574 154839 '

14 5/$23375 24/$74040 1165/$21574 118989 15 5/$23375 26/$80210 1165/$21574 l

125159 16 [

5/$23375 24/$74040 1165/$21574 118989 17 3/$14025 f

25/$77125 828/$15333 106483  !

18 3/$14025 25/$77125 1165/$21574

?

112724 [

19 3/$14025 25/$77125  !

920/$17037 108187 '

20 3/$14025 27/$83295 +

920/$17037 114357 21 3/$14025 25/$7,125 t

980/$18148 109298 (

22 27/$77125  !

980/$18148 95273 '

23 i 27/$39960 1250/$23148 63108' 24 27/$39960 1250/$23148 63108 25 27/$39960 1250/$23148 63108 l

17

i References ,

1. R.F. Barth, A.H. Soloway, R.G. Fairchild. Boron Neutron Capture Therapy of Cancer. Cancer Research 1990; 50:1061-70. ,

2. R.F. Barth, A.H. Soloway, R.G. Fairchild. Boron Neutron Capture Therapy of Cancer. Scientific American, Oct., 1990. ,

+

3. R.F. Barth, A.H. Soloway, R.G. Fairchild. Boron Neutron Capture  !

Therapy of Cancer. Cancer 1992; 70:2995-3007.

4. J.H. Morris. Boron Neutron Capture Therapy. Chemistry in .

Britain, 1991; 27:331-34.

5. Development and Deployment of a Revolutionary Brain Tumor Treatment. Boron Neutron Capture Therapy - University Consortium, Inc. 1994 i

6. T. Bays. Boron Neutron Capture Therapy for Cancer. 1994

7. F.J. Wheeler, D.W. Nigg. Three dimensional radiation dose f distributio, analysis for boron neutron capture therapy. Nuclear Sci 6 Engineering 1992; 110:16-31.

8. R.L. Moss, R.A. Stecher, F. Asmussen, R. Huiskamp, L. Dewit, Mijnheer. The Petten BNCT project. In: B.G. Allen, D. Moore, B. B.

Harrington, editors. Progress in neutron capture therapy for i cancer. New York: Plenum Press, 1992 (In press).

9. F.J. Wheeler, D.W. Nigg. Feasibility Study for an Epithermal I Neutron-Beam Facility at the Washington State University Radiation Center. EGG-NRE-11296, 1994.

10. B.H. Liu et al. Enhancement of the Epithermal Neutron Beam Used for Boron Neutron Capture Therapy. Int. J. Radiation Oncology Biol.

Phys., 1994; 28:1149-1156. .!

i t

9 h

i 18

?

-- -