Development of a three-dimensionally movable phantom system for dosimetric verifications Hiroshi Nakayama Department of Radiation Oncology and Image-Applied Therapy, Graduate School of Medicine, Kyoto University, 54 Shogoin-Kawaharacho, Sakyo-ku, Kyoto, 606-8507, Japan and Medical Systems Administration Office, Hiroshima Machinery Works, Mitsubishi Heavy Industries, Limited, 4-6-22, Kan-On-Shin-Machi, Nishi-Ku, Hiroshima, 733-8553, Japan Takashi Mizowakia and Yuichiro Narita Department of Radiation Oncology and Image-Applied Therapy, Graduate School of Medicine, Kyoto University, 54 Shogoin-Kawaharacho, Sakyo-ku, Kyoto, 606-8507, Japan Noriyuki Kawada, Kunio Takahashi, and Kazumasa Mihara Medical Systems Administration Office, Hiroshima Machinery Works, Mitsubishi Heavy Industries, Limited, 4-6-22, Kan-On-Shin-Machi, Nishi-Ku, Hiroshima, 733-8553, Japan Masahiro Hiraoka Department of Radiation Oncology and Image-Applied Therapy, Graduate School of Medicine, Kyoto University, 54 Shogoin-Kawaharacho, Sakyo-ku, Kyoto, 606-8507, Japan Received 30 August 2007 revised 19 February 2008 accepted for publication 19 February 2008 published 7 April 2008 The authors developed a three-dimensionally movable phantom system 3D movable phantom system which can reproduce three-dimensional movements to experimentally verify the impact of radiotherapy treatment-related movements on dose distribution. The phantom system consists of three integrated components: a three-dimensional driving mechanism 3D driving mechanism , computer control system, and phantoms for film dosimetry. The 3D driving mechanism is a quint- essential part of this system. It is composed of three linear-motion tables single-axis robots which are joined orthogonally to each other. This mechanism has a motion range of 100 mm, with a maximum velocity of 200 mm/s in each dimension, and 3D motion ability of arbitrary patterns. These attributes are sufficient to reproduce almost all organ movements. The positional accuracy of this 3D movable phantom system in a state of geostationary is less than 0.1 mm. The maximum error in terms of the absolute position on movement was 0.56 mm. The positional reappearance error on movement was up to 0.23 mm. The observed fluctuation of time was 0.012 s in the cycle of 4.5 s of oscillation. These results suggested that the 3D movable phantom system exhibited a sufficient level of accuracy in terms of geometry and timing to reproduce interfractional organ movement or setup errors in order to assess the influence of these errors on high-precision radio- therapy such as stereotactic irradiation and intensity-modulated radiotherapy. In addition, the au- thors 3D movable phantom system will also be useful in evaluating the adequacy and efficacy of new treatment techniques such as gating or tracking radiotherapy. �� 2008 American Association of Physicists in Medicine. DOI: 10.1118/1.2897971 Key words: organ motion, simulation, quality assurance, radiation therapy I. INTRODUCTION Recently, high-precision radiotherapy, such as stereotactic ir- radiation and intensity-modulated radiotherapy IMRT , has become widely applied in routine clinical practices through- out the world. With these techniques, the delivery of excel- lent dose concentrations to targets and the sparing of organs at risk OARs can be achievable along with a steep dose gradient around the targets. Furthermore, in IMRT, nonuni- form photon fluence in a field is achieved by changing field shapes as the irradiation time elapses.1,2 Because of the fea- tures of high-precision radiotherapy described above, it is expected that treatment-related errors such as setup errors and the errors occurred by internal organ movements can lead to a much larger impact on the dose actually delivered to the targets and OARs compared to traditional irradiation techniques with relatively larger margins and uniform beam fluence in each radiation field. In other words, even a slight positional divergence of targets or OARs from a planned position causes large differences between planned and actu- ally delivered dose distributions.3���5 The existing software commercially available for radiotherapy treatment planning RTP , however, cannot take into account such uncertainties in planning treatment and calculating dose. Treatment plans therefore are currently created using an image data set of static computed tomography CT scans as a standard clini- cal practice and evaluated based on the presumption that ob- jects do not move at all, and only static dose distributions are measured by chamber or film dosimetry for quality assurance of IMRT.6,7 To overcome the above-described problems, a few movable phantom systems, which facilitate the evalua- 1643 1643 Med. Phys. 35 ���5���, May 2008 0094-2405/2008/35���5���/1643/8/$23.00 �� 2008 Am. Assoc. Phys. Med.

tion of impacts on treatment-related errors of dose distribu- tion, have been introduced.8���10 These systems, however, have not been able to completely realize the arbitrary three- dimensional 3D movements.8���10 Other systems, which re- produce arbitrary movements have been developed, however, no system of them has been achieved the sufficient positional accuracy for the quality assurance of IMRT under the condi- tion of driving a heavy dosimetric phantom.11���13 We there- fore developed the 3D movable phantom system that can load heavy dosimetric phantoms and accurately simulate ar- bitrary 3D movements. In the present article, details of the phantom system are described, and the accuracy of the sys- tem in simulating 3D movements is evaluated. II. MATERIALS AND METHODS II.A. 3D movable phantom system II.A.1. System overview The 3D movable phantom system is designed to fit linear accelerators or CT. This system consists of the following three components: a 3D driving mechanism, a computers control system, and phantoms for film dosimetry. Configura- tions of the system are indicated in Fig. 1. II.A.2. 3D phantom driving mechanism The 3D driving mechanism is composed of a phantom support base and three linear actuators along with three dif- ferent axes, those are called single-axis robots. The single- axis robot is a device with a slider that can move on a linear guide fixed to the outer frame. The slider is able to move arbitrarily on the linear guide by a ball-screw mechanism. The first single-axis robot is fixed to an iron base plate in a direction whereby the slider of the robot is able to move, along the cephalocaudal axis Y axis . This robot is called the Y-axis robot. The second robot, which is called the X-axis robot, is fixed to the slider of the Y-axis robot, and the slider of the X-axis robot can move in a left-and-right direction X axis . The third and last robot, called the Z-axis robot, is fixed to the slider of the X-axis robot and can move parallel to the vertical direction Z axis . The support base for the phantoms is mounted on the slider of the Z-axis robot Fig. 1 . The X- and Y-axis robots are ISA-SXM-1-60-4-100- T1-AQ IAI Corp., Shizuoka, Japan , and the Z-axis robot is ISA-SXM-1-60-4-100-T1-AQ-B-NM IAI Corp., Shizuoka, Japan , which is a model added a braking mechanism to the foregoing. Details of the performance specifications of these robots are indicated in Table I.27 The dimensions of the 3D driving mechanism with the iron base plate are: 41, 34, and 38 cm in width X axially , length Y axially , and height Z axially , respectively. The total weight is 23.6 kg. II.A.3. Computers control system The 3D driving mechanism is composed only of mechani- cal parts except for a positional sensor and a driving motor. ! #$%&'() *+,-,&. )/0$%&,1) %&* #$%&'() 2(+ *(1,)/'+3 4()#5'/+ 0(&'+(6 131'/) 7849 0(&'+(66/+9 %&* 1(2':%+/ +=,'+%+3 ! )(',(& *%'% 4(&'+(6 %&* %0'5%6 )(-,&. *%'% %05,1,',( ? @A % , 1 B @ % A , 1 C@%A,1 8$%&'() FIG. 1. 3D movable phantom system. The photo shows the 3D driving mechanism and attached dosimetric phantom. TABLE I. Performance specification of the single-axis robot. Model ISA-SXM-1-60-4-100-T1-AQ or ISA-SXM-1-60-4-100- T1-AQ-B-NM Movable range 100 mm Maximum acceleration Horizontal 0.5 G 1.0 G=9800 mm / s2 Vertical 0.3 G 1.0 G=9800 mm / s2 Load-bearing capacity Horizontal 30 kg Vertical 12 kg Maximum speed 200 mm/s Repetition positioning accuracy 0.02 mm Drive system 12 mm diameter ball screw rolled thread C10 Backlash 0.05 mm or less Usable temperature range From 0 to 40 ��C 1644 Nakayama et al.: Three-dimensionally movable phantom system 1644 Medical Physics, Vol. 35, No. 5, May 2008