Development of twist‐correction system for radiotherapy of head and neck cancer patients

Abstract To propose a concept for correcting the twist between the head and neck and the body frequently occurring in radiotherapy patients and to develop a prototype device for achieving this. Furthermore, the operational accuracy of this device under no load was evaluated. We devised a concept for correcting the twist of patients by adjustment of the three rotation (pitch, roll, and yaw) angles in two independent plates connected by a joint (fulcrum). The two plates (head and neck plate and body plate) rotate around the fulcrum by adjusting screws under each of them. A prototype device was created to materialize this concept. First, after all adjusting screws were set to the zero position, the rotation angle of each plate was measured by a digital goniometer. Repeatability was evaluated by performing 20 repeated measurements. Next, to confirm the rotational accuracy of each plate of the prototype device, the calculated rotation angles for 20 combinations of patterns of traveled distances of the adjusting screws were compared with those measured by the digital goniometer and cone‐beam computed tomography (CT). The repeatability (standard deviation: SD) of the pitch, roll, and yaw angles of the head and neck plate was 0.04°, 0.05°, and 0.03°, and the repeatability (SD) of the body plate was 0.05°, 0.04°, and 0.04°, respectively. The mean differences ± SD between the calculated and measured pitch, roll, and yaw angles for the head and neck plate with the digital goniometer were 0.00 ± 0.06°, −0.01 ± 0.06°, and −0.04 ± 0.04°, respectively. The differences for the body plate were −0.03 ± 0.04°, 0.03 ± 0.05°, and 0.02 ± 0.05°, respectively. Results of the cone‐beam CT were similar to those of the digital goniometer. The prototype device exhibited good performance regarding the rotational accuracy and repeatability under no load. The clinical implementation of this concept is expected to reduce the residual error of the patient position due to the twist.


| INTRODUCTION
Intensity-modulated radiotherapy (IMRT) can deliver the dose to tumor lesions locally and reduce the dose delivered to the normal tissue surrounding the tumor. Dose escalation to the tumor by this technique can improve the treatment outcome. In particular, IMRT for cases of head and neck (HN) cancer can deliver adequate doses to complex-shaped tumors surrounding normal tissues, such as the spinal cord and brain stem. In addition, IMRT can reduce the dose to the parotid glands, submandibular glands, and oral cavity; therefore, it alleviates salivary disorder and improves the patient's quality of life. On the other hand, to maximize these advantages of IMRT, it is important to use an image-guided system to match the patient's position in the radiotherapy with that in the treatment planning computed tomography (CT). However, we often experience misalignment of the tumor shape due to twisting of the patient's neck. Correction by the current image-guided system can be applied only to six axes (three translation axes and three rotation axes) to the whole body of the patient, 1-3 so the partial twist of the patient cannot be completely canceled. [4][5][6][7] This results in errors remaining even after correction. Such potential error is compensated for by adding a safety margin around the tumor; however, this approach interferes with the benefits of IMRT, which can deliver the dose to tumor lesions locally. The aim of this study was to propose a concept for correcting the above-mentioned twist of the patient and to develop a prototype device for achieving this. Furthermore, the operational accuracy of this device under no load was evaluated.

2.A | Schema of the prototype
We devised a concept that can correct the twist of the patient by adjustment of the three rotation angles in two independent plates connected by a joint (fulcrum). Figure 1 shows the prototype device made to materialize this concept. As illustrated in Fig. 1 Table 1.

2.B | Theory of rotation angle calculation
Each plate rotates around the fulcrum using three adjusting screws under the plate. Figure 3 shows the relationship between the rotation angle and the traveled distance of each adjusting screw. For example, the pitch and roll angles (the rotation angles for the x-and y-axes, respectively) of the body plate can be obtained by adjusting  Fig. 3(c). The rotation angle with respect to each translation axis (x, y, and z) of the plate is calculated by the traveled distance of the adjusting screw and the known distance between the adjusting screws as follows: where θ i pitch , θ i roll , and θ i yaw show the pitch, roll, and yaw angles of plate i (HN or body) and d indicates the traveled distance of the adjusting screw. When i is HN, d(i p ) shows the traveled distance of the adjusting screw HN S , and when i is body, d(i p ) shows that of the adjusting screw B I . The movable ranges of θ i pitch , θ i roll , and θ i yaw calculated from the traveled distance of the adjusting screw were from −2.66 to 2.66, −5.08 to 5.08, and −2.39 to 2.39, respectively. The clockwise direction for each translation axis was defined as positive angles.

2.C | Operational accuracy of the prototype device
First, after all adjusting screws were set to the zero position, the rotation angle of each plate was measured by a digital goniometer (DP-90; Niigata Seiki Co., Ltd.; detection limit: 0.05°). Repeatability was evaluated by 20 repeated measurements.
Next, 20 combinations of traveled distances of the six adjusting screws were obtained using random numbers from Microsoft Excel 2010 (Table 2). To confirm the rotational accuracy of the prototype device, θ i pitch , θ i roll , and θ i yaw calculated at a given traveled distance were compared with those measured by a digital goniometer. Since it was not possible to measure the yaw angle when the device was lying on the floor, this measurement was performed at a position of 90°to the floor. In addition, ten acrylic cubic phantoms with tungsten sphere of 1 mm in diameter were placed on the prototype device (Fig. 4), and after movement by 20 combinations of six adjustment screws, scanning was performed with a Varian conebeam CT imaging system. The resolutions of the cone-beam CT images for x, y, and z directions were 0.5, 1.0, and 0.5 mm,   Table 3 shows the values of θ i pitch , θ i roll , and θ i yaw calculated by the combination patterns for the traveled distances of the adjusting screws in Table 2. Figure 5 shows the differences between the calculated and measured rotation angles in the combinations of 20 patterns for the adjusting screws. The mean differences ± SD of θ HN pitch , θ HN roll , and θ HN yaw by digital goniometer were 0.00 ± 0.06°, −0.01 ± 0.06°, and −0.04 ± 0.04°, respectively. The differences of θ body pitch , θ body roll , and θ body yaw were −0.03 ± 0.04°, 0.03 ± 0.05°, and 0.02 ± 0.05°, respectively. The mean differences ± SD of θ HN pitch , θ HN roll , and θ HN yaw by cone-beam CT were 0.06 ± 0.06°, 0.00 ± 0.03°, and 0.04 ± 0.06°, respectively. The differences of θ body pitch , θ body roll , and θ body yaw were 0.00 ± 0.06°, 0.04 ± 0.08°, and 0.00 ± 0.04°, respectively.

| DISCUSSION
In this paper, we propose a concept for correcting the twist that is frequently exhibited in the neck region of patients undergoing radiotherapy and developed a prototype device for embodying this concept. To the best of our knowledge, this is the first report of a device applied for correcting the twist of a patient. The prototype device might be able to reduce the residual error of the patient position due to this twist and reduce the additional margin needed for compensating the adequate dose to tumors.
Correction of the patient position by the current image-guided system is mainly performed on six axes using the six degrees of freedom (6DoF) couch. There is also a treatment machine, called tomotherapy, which can correct only three translational and one rotational axes. 8 Zhang et al. investigated the rotational accuracy of the 6DoF couch with cone-beam CT, 9 and they reported that the mean rotational errors ± SD for pitch, roll, and yaw angles were 0.028 ± 0.032°, −0.043 ± 0.058°, and −0.009 ± 0.033°, respectively.
The rotational accuracy of our prototype device was measured using  tories. 11,12 These systems can correct the rotational position of the patient by themselves; therefore, they might be able to correct the twist with the couch, which has the ability to achieve rotational correction. Our system has versatility because it can add twist-correction to a radiotherapy device with a couch that can only correct translation movement.
The density of polyoxymethylene and polyether ether ketone materials in Table 1 was smaller than that of the cortical bone material (Gammex, Inc., density; 1.559 g cm −3 ). RENY with the nearest density to that of the cortical bone was used in the screw for connecting the parts of the prototype device, but the amount used in the prototype device was small. There is no influence on the target delineation because there are few artifacts in the CT images. If the  Next, our device might not be suitable for cases with sufficient safety margin for the tumor. The efficacy of the device would be shown for the cases, wherein the accuracy of patient position is particularly required (e.g., IMRT for HN cancer and stereotactic radiotherapy for the recurrence). Finally, our device can be attached to the CT and treatment couches with commercial lock bars; therefore, the position reproducibility of the device is high. Conversely, the flatness of the patient needs to be maintained by setting a cushion on the patient's back to fill the difference in level between the couch and the body plate of the device.

| CONCLUSION S
We developed a twist-correction system for HN radiotherapy. We found that the rotational accuracy and repeatability exhibited good performance under no load. Our prototype device still has some challenges for a clinical application: the development of special software for calculating the rotational angles and equipment for automatically correcting the angles; the mechanism for attaching the thermoplastic mask on the HN and body plates. Conversely, the clinical application of this concept could be expected to significantly reduce the residual F I G . 5. The differences between the calculated and measured rotation angles in the combinations of 20 patterns for the adjusting screws: (a) for the HN plate with the digital goniometer, (b) for the body plate with the digital goniometer, (c) for the HN plate with cone-beam computed tomography, and (d) for the body plate with the cone-beam computed tomography