Toshiyuki Terunuma
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Toshiyuki Terunuma has a total of 221 co-authors. Here are the top co-authors by number of shared publications:
Takeji Sakae
Institute of Medicine University of Tsukuba Ibaraki Japan
30 shared publicationsAkihiro Nohtomi
Atomic Energy Research Institute , Kinki University , 3-4-1 Kowakae, Higashiosaka-shi, Osaka 577-8502 , Japan
14 shared publicationsHideyuki Sakurai
Institute of Medicine University of Tsukuba Ibaraki Japan
9 shared publications- Takeji Sakae
Proton Medical Research Center, University of Tsukuba, Ibaraki, Japan
9 shared publications Kiyoshi Yasuoka
8 shared publicationsRyosuke Kohno
8 shared publications
Latest Publications
- Open Access
- 8 November 2025
- Published by Wiley in Medical Physics
- Vol. 52 (11) , e70121
Characterization of secondary‐radiation background in X‐ray flat‐panel detectors during scanning proton beam irradiation
Background: Secondary‐radiation background (BG) generated in kilovoltage X‐ray flat‐panel detectors (FPDs) under scanning proton beam irradiation has not been thoroughly analyzed. Purpose: This study aimed to determine and mathematically model the characteristics of secondary‐radiation BG in FPDs under scanning proton beam irradiation. Methods: Using a synchrotron‐based proton system and two FPDs mounted on a gantry tilted at 135° to the proton beam axis, we acquired images of two phantoms (block and thorax). The FPD images were captured during three conditions: only X‐ray exposure, proton irradiation, and pauses between proton irradiations. Because the FPD readout rate (30 fps) was twice the X‐ray exposure rate (15 pps), the images were further categorized into two types: those corresponding to with and without X‐ray exposure. These FPD images were analyzed to determine the characteristics of the secondary‐radiation BG. In addition, a Monte Carlo simulation was conducted to complement the performed measurements. Results: Analysis of the FPD images without X‐ray exposure revealed that the pure secondary‐radiation BG appeared as sparse spike patterns without a noticeable spatial bias across the FPD area in most images. The BG affected only 1.25% of the FPD pixels and could be modeled as an exponentially decreasing function with increasing pixel intensity. When considering typical X‐ray image brightness for the block and thorax phantoms, the impact of the BG on mean‐pixel‐intensity variation was < 1%. Monte Carlo simulations suggested that prompt photons were the primary source reaching the FPD. Conclusions: This study demonstrates that the secondary‐radiation BG characteristics observed in the X‐ray FPD images during scanning proton beam irradiation can be quantitatively modeled. The impact of the secondary‐radiation BG on mean‐pixel‐intensity of FPD images was minimal.
- 8 November 2025
- Published by Wiley in Medical Physics
- Vol. 52 (11) , e70121
Open AccessCharacterization of secondary‐radiation background in X‐ray flat‐panel detectors during scanning proton beam irradiation
Background: Secondary‐radiation background (BG) generated in kilovoltage X‐ray flat‐panel detectors (FPDs) under scanning proton beam irradiation has not been thoroughly analyzed. Purpose: This study aimed to determine and mathematically model the characteristics of secondary‐radiation BG in FPDs under scanning proton beam irradiation. Methods: Using a synchrotron‐based proton system and two FPDs mounted on a gantry tilted at 135° to the proton beam axis, we acquired images of two phantoms (block and thorax). The FPD images were captured during three conditions: only X‐ray exposure, proton irradiation, and pauses between proton irradiations. Because the FPD readout rate (30 fps) was twice the X‐ray exposure rate (15 pps), the images were further categorized into two types: those corresponding to with and without X‐ray exposure. These FPD images were analyzed to determine the characteristics of the secondary‐radiation BG. In addition, a Monte Carlo simulation was conducted to complement the performed measurements. Results: Analysis of the FPD images without X‐ray exposure revealed that the pure secondary‐radiation BG appeared as sparse spike patterns without a noticeable spatial bias across the FPD area in most images. The BG affected only 1.25% of the FPD pixels and could be modeled as an exponentially decreasing function with increasing pixel intensity. When considering typical X‐ray image brightness for the block and thorax phantoms, the impact of the BG on mean‐pixel‐intensity variation was < 1%. Monte Carlo simulations suggested that prompt photons were the primary source reaching the FPD. Conclusions: This study demonstrates that the secondary‐radiation BG characteristics observed in the X‐ray FPD images during scanning proton beam irradiation can be quantitatively modeled. The impact of the secondary‐radiation BG on mean‐pixel‐intensity of FPD images was minimal.
- Open Access
- 1 October 2025
- Published by Wolters Kluwer Health in Journal of Medical Physics
- Vol. 50 (4) , 636-649
Uncertain Feature-refinement Attention Unet: Considering Suitable Convolutional Neural Network Model for Real-time Segmentation in Markerless Tumor Tracking
Purpose: In markerless tumor tracking (MTT) with deep learning, model performance suffers from domain shifts due to noise and anatomical changes. This study aimed to develop a convolutional neural network (CNN) model for real-time MTT segmentation. Methods: An Uncertain Feature-refinement Attention Unet (UFA-Unet), designed based on insights into CNN behavior under domain distribution shifts that occur between digitally reconstructed radiographs (DRRs) and kV X-ray fluoroscopic (XF) images, is proposed. A qualitative ablation study was performed to examine the contribution of each UFA-Unet component to segmentation accuracy. The model feasibility of UFA-Unet was evaluated through quantitative and phantom studies. The quantitative study included ten lung cancer cases, each containing two datasets (1 st -plan and 2 nd -plan), with a mean interval of 28 days between four-dimensional computed tomography (4DCT) acquisitions. Patient-specific models were trained on 1 st -plan DRRs and validated using noise-injected 1 st -and 2 nd -plan DRRs. In the phantom study, UFA-Unet was trained with only a single exhalation phase (T50) of 4DCT data and evaluated using dynamic phantom XF images with 25-mm amplitude motion. UFA-Unet was compared against U-Net, Attention-Unet, and Swin-Unet. Results: The ablation study confirmed that each component suppressed over-activation to improve segmentation accuracy. In the quantitative study, UFA-Unet maintained superior performance compared with conventional models on both 1 st - and 2 nd -plan DRRs with noise injection. Furthermore, in the phantom study, UFA-Unet demonstrated robust tracking under previously unseen respiratory phases, achieving a 95 th percentile 3D error of 0.61–3.13 mm and consistently outperforming conventional models. Conclusion: UFA-Unet provides accurate, robust, and real-time segmentation, thus demonstrating its suitability for clinical MTT.
- 1 October 2025
- Published by Wolters Kluwer Health in Journal of Medical Physics
- Vol. 50 (4) , 636-649
Open AccessUncertain Feature-refinement Attention Unet: Considering Suitable Convolutional Neural Network Model for Real-time Segmentation in Markerless Tumor Tracking
Purpose: In markerless tumor tracking (MTT) with deep learning, model performance suffers from domain shifts due to noise and anatomical changes. This study aimed to develop a convolutional neural network (CNN) model for real-time MTT segmentation. Methods: An Uncertain Feature-refinement Attention Unet (UFA-Unet), designed based on insights into CNN behavior under domain distribution shifts that occur between digitally reconstructed radiographs (DRRs) and kV X-ray fluoroscopic (XF) images, is proposed. A qualitative ablation study was performed to examine the contribution of each UFA-Unet component to segmentation accuracy. The model feasibility of UFA-Unet was evaluated through quantitative and phantom studies. The quantitative study included ten lung cancer cases, each containing two datasets (1 st -plan and 2 nd -plan), with a mean interval of 28 days between four-dimensional computed tomography (4DCT) acquisitions. Patient-specific models were trained on 1 st -plan DRRs and validated using noise-injected 1 st -and 2 nd -plan DRRs. In the phantom study, UFA-Unet was trained with only a single exhalation phase (T50) of 4DCT data and evaluated using dynamic phantom XF images with 25-mm amplitude motion. UFA-Unet was compared against U-Net, Attention-Unet, and Swin-Unet. Results: The ablation study confirmed that each component suppressed over-activation to improve segmentation accuracy. In the quantitative study, UFA-Unet maintained superior performance compared with conventional models on both 1 st - and 2 nd -plan DRRs with noise injection. Furthermore, in the phantom study, UFA-Unet demonstrated robust tracking under previously unseen respiratory phases, achieving a 95 th percentile 3D error of 0.61–3.13 mm and consistently outperforming conventional models. Conclusion: UFA-Unet provides accurate, robust, and real-time segmentation, thus demonstrating its suitability for clinical MTT.
- Open Access
- 24 November 2022
- Published by Wiley in Medical Physics
- Vol. 50 (1) , 480-494
Explainability and controllability of patient‐specific deep learning with attention‐based augmentation for markerless image‐guided radiotherapy
Background: We reported the concept of patient‐specific deep learning (DL) for real‐time markerless tumor segmentation in image‐guided radiotherapy (IGRT). The method was aimed to control the attention of convolutional neural networks (CNNs) by artificial differences in co‐occurrence probability (CoOCP) in training datasets, that is, focusing CNN attention on soft tissues while ignoring bones. However, the effectiveness of this attention‐based data augmentation has not been confirmed by explainable techniques. Furthermore, compared to reasonable ground truths, the feasibility of tumor segmentation in clinical kilovolt (kV) X‐ray fluoroscopic (XF) images has not been confirmed.Purpose: The first aim of this paper was to present evidence that the proposed method provides an explanation and control of DL behavior. The second purpose was to validate the real‐time lung tumor segmentation in clinical kV XF images for IGRT.Methods: This retrospective study included 10 patients with lung cancer. Patient‐specific and XF angle‐specific image pairs comprising digitally reconstructed radiographs (DRRs) and projected‐clinical‐target‐volume (pCTV) images were calculated from four‐dimensional computer tomographic data and treatment planning information. The training datasets were primarily augmented by random overlay (RO) and noise injection (NI): RO aims to differentiate positional CoOCP in soft tissues and bones, and NI aims to make a difference in the frequency of occurrence of local and global image features. The CNNs for each patient‐and‐angle were automatically optimized in the DL training stage to transform the training DRRs into pCTV images. In the inference stage, the trained CNNs transformed the test XF images into pCTV images, thus identifying target positions and shapes.Results: The visual analysis of DL attention heatmaps for a test image demonstrated that our method focused CNN attention on soft tissue and global image features rather than bones and local features. The processing time for each patient‐and‐angle‐specific dataset in the training stage was ∼30 min, whereas that in the inference stage was 8 ms/frame. The estimated three‐dimensional 95 percentile tracking error, Jaccard index, and Hausdorff distance for 10 patients were 1.3–3.9 mm, 0.85–0.94, and 0.6–4.9 mm, respectively.Conclusions: The proposed attention‐based data augmentation with both RO and NI made the CNN behavior more explainable and more controllable. The results obtained demonstrated the feasibility of real‐time markerless lung tumor segmentation in kV XF images for IGRT.
- 24 November 2022
- Published by Wiley in Medical Physics
- Vol. 50 (1) , 480-494
Open AccessExplainability and controllability of patient‐specific deep learning with attention‐based augmentation for markerless image‐guided radiotherapy
Background: We reported the concept of patient‐specific deep learning (DL) for real‐time markerless tumor segmentation in image‐guided radiotherapy (IGRT). The method was aimed to control the attention of convolutional neural networks (CNNs) by artificial differences in co‐occurrence probability (CoOCP) in training datasets, that is, focusing CNN attention on soft tissues while ignoring bones. However, the effectiveness of this attention‐based data augmentation has not been confirmed by explainable techniques. Furthermore, compared to reasonable ground truths, the feasibility of tumor segmentation in clinical kilovolt (kV) X‐ray fluoroscopic (XF) images has not been confirmed.Purpose: The first aim of this paper was to present evidence that the proposed method provides an explanation and control of DL behavior. The second purpose was to validate the real‐time lung tumor segmentation in clinical kV XF images for IGRT.Methods: This retrospective study included 10 patients with lung cancer. Patient‐specific and XF angle‐specific image pairs comprising digitally reconstructed radiographs (DRRs) and projected‐clinical‐target‐volume (pCTV) images were calculated from four‐dimensional computer tomographic data and treatment planning information. The training datasets were primarily augmented by random overlay (RO) and noise injection (NI): RO aims to differentiate positional CoOCP in soft tissues and bones, and NI aims to make a difference in the frequency of occurrence of local and global image features. The CNNs for each patient‐and‐angle were automatically optimized in the DL training stage to transform the training DRRs into pCTV images. In the inference stage, the trained CNNs transformed the test XF images into pCTV images, thus identifying target positions and shapes.Results: The visual analysis of DL attention heatmaps for a test image demonstrated that our method focused CNN attention on soft tissue and global image features rather than bones and local features. The processing time for each patient‐and‐angle‐specific dataset in the training stage was ∼30 min, whereas that in the inference stage was 8 ms/frame. The estimated three‐dimensional 95 percentile tracking error, Jaccard index, and Hausdorff distance for 10 patients were 1.3–3.9 mm, 0.85–0.94, and 0.6–4.9 mm, respectively.Conclusions: The proposed attention‐based data augmentation with both RO and NI made the CNN behavior more explainable and more controllable. The results obtained demonstrated the feasibility of real‐time markerless lung tumor segmentation in kV XF images for IGRT.
- 25 January 2020
- Published by Springer Nature
- p. 127-138
Motion Management
- Open Access
- 12 August 2018
- Published by Springer Nature in Radiological Physics and Technology
- Vol. 11 (3) , 362-363
Response to “Comments on ‘Novel real-time tumor-contouring method using deep learning to prevent mistracking in X-ray fluoroscopy”’
- 12 August 2018
- Published by Springer Nature in Radiological Physics and Technology
- Vol. 11 (3) , 362-363
Open AccessResponse to “Comments on ‘Novel real-time tumor-contouring method using deep learning to prevent mistracking in X-ray fluoroscopy”’
- Open Access
- 28 December 2017
- Published by Springer Nature in Radiological Physics and Technology
- Vol. 11 (1) , 43-53
Novel real-time tumor-contouring method using deep learning to prevent mistracking in X-ray fluoroscopy
Robustness to obstacles is the most important factor necessary to achieve accurate tumor tracking without fiducial markers. Some high-density structures, such as bone, are enhanced on X-ray fluoroscopic images, which cause tumor mistracking. Tumor tracking should be performed by controlling “importance recognition”: the understanding that soft-tissue is an important tracking feature and bone structure is unimportant. We propose a new real-time tumor-contouring method that uses deep learning with importance recognition control. The novelty of the proposed method is the combination of the devised random overlay method and supervised deep learning to induce the recognition of structures in tumor contouring as important or unimportant. This method can be used for tumor contouring because it uses deep learning to perform image segmentation. Our results from a simulated fluoroscopy model showed accurate tracking of a low-visibility tumor with an error of approximately 1 mm, even if enhanced bone structure acted as an obstacle. A high similarity of approximately 0.95 on the Jaccard index was observed between the segmented and ground truth tumor regions. A short processing time of 25 ms was achieved. The results of this simulated fluoroscopy model support the feasibility of robust real-time tumor contouring with fluoroscopy. Further studies using clinical fluoroscopy are highly anticipated.
- 28 December 2017
- Published by Springer Nature in Radiological Physics and Technology
- Vol. 11 (1) , 43-53
Open AccessNovel real-time tumor-contouring method using deep learning to prevent mistracking in X-ray fluoroscopy
Robustness to obstacles is the most important factor necessary to achieve accurate tumor tracking without fiducial markers. Some high-density structures, such as bone, are enhanced on X-ray fluoroscopic images, which cause tumor mistracking. Tumor tracking should be performed by controlling “importance recognition”: the understanding that soft-tissue is an important tracking feature and bone structure is unimportant. We propose a new real-time tumor-contouring method that uses deep learning with importance recognition control. The novelty of the proposed method is the combination of the devised random overlay method and supervised deep learning to induce the recognition of structures in tumor contouring as important or unimportant. This method can be used for tumor contouring because it uses deep learning to perform image segmentation. Our results from a simulated fluoroscopy model showed accurate tracking of a low-visibility tumor with an error of approximately 1 mm, even if enhanced bone structure acted as an obstacle. A high similarity of approximately 0.95 on the Jaccard index was observed between the segmented and ground truth tumor regions. A short processing time of 25 ms was achieved. The results of this simulated fluoroscopy model support the feasibility of robust real-time tumor contouring with fluoroscopy. Further studies using clinical fluoroscopy are highly anticipated.
- Open Access
- 4 May 2017
- Published by Springer Nature in BMC Medical Imaging
- Vol. 17 (1) , 30
Registration error of the liver CT using deformable image registration of MIM Maestro and Velocity AI
Understanding the irradiated area and dose correctly is important for the reirradiation of organs that deform after irradiation, such as the liver. We investigated the spatial registration error using the deformable image registration (DIR) software products MIM Maestro (MIM) and Velocity AI (Velocity). Image registration of pretreatment computed tomography (CT) and posttreatment CT was performed in 24 patients with liver tumors. All the patients received proton beam therapy, and the follow-up period was 4–14 (median: 10) months. We performed DIR of the pretreatment CT and compared it with that of the posttreatment CT by calculating the dislocation of metallic markers (implanted close to the tumors). The fiducial registration error was comparable in both products: 0.4–32.9 (9.3 ± 9.9) mm for MIM and 0.5–38.6 (11.0 ± 10.0) mm for Velocity, and correlated with the tumor diameter for MIM (r = 0.69, P = 0.002) and for Velocity (r = 0.68, P = 0.0003). Regarding the enhancement effect, the fiducial registration error was 1.0–24.9 (7.4 ± 7.7) mm for MIM and 0.3–29.6 (8.9 ± 7.2) mm for Velocity, which is shorter than that of plain CT (P = 0.04, for both). The DIR performance of both MIM and Velocity is comparable with regard to the liver. The fiducial registration error of DIR depends on the tumor diameter. Furthermore, contrast-enhanced CT improves the accuracy of both MIM and Velocity. H28-102; July 14, 2016 approved.
- 4 May 2017
- Published by Springer Nature in BMC Medical Imaging
- Vol. 17 (1) , 30
Open AccessRegistration error of the liver CT using deformable image registration of MIM Maestro and Velocity AI
Understanding the irradiated area and dose correctly is important for the reirradiation of organs that deform after irradiation, such as the liver. We investigated the spatial registration error using the deformable image registration (DIR) software products MIM Maestro (MIM) and Velocity AI (Velocity). Image registration of pretreatment computed tomography (CT) and posttreatment CT was performed in 24 patients with liver tumors. All the patients received proton beam therapy, and the follow-up period was 4–14 (median: 10) months. We performed DIR of the pretreatment CT and compared it with that of the posttreatment CT by calculating the dislocation of metallic markers (implanted close to the tumors). The fiducial registration error was comparable in both products: 0.4–32.9 (9.3 ± 9.9) mm for MIM and 0.5–38.6 (11.0 ± 10.0) mm for Velocity, and correlated with the tumor diameter for MIM (r = 0.69, P = 0.002) and for Velocity (r = 0.68, P = 0.0003). Regarding the enhancement effect, the fiducial registration error was 1.0–24.9 (7.4 ± 7.7) mm for MIM and 0.3–29.6 (8.9 ± 7.2) mm for Velocity, which is shorter than that of plain CT (P = 0.04, for both). The DIR performance of both MIM and Velocity is comparable with regard to the liver. The fiducial registration error of DIR depends on the tumor diameter. Furthermore, contrast-enhanced CT improves the accuracy of both MIM and Velocity. H28-102; July 14, 2016 approved.
- Open Access
- 22 February 2017
- Published by Springer Nature in Radiation Oncology
- Vol. 12 (1) , 40
A comparative study of dose distribution of PBT, 3D-CRT and IMRT for pediatric brain tumors
It was reported that proton beam therapy (PBT) reduced the normal brain dose compared with X-ray therapy for pediatric brain tumors. We considered whether there was not the condition that PBT was more disadvantageous than intensity modulated photon radiotherapy (IMRT) and 3D conventional radiotherapy (3D-CRT) for treatment of pediatric brain tumors about the dose reduction for the normal brain when the tumor location or tumor size were different. The subjects were 12 patients treated with PBT at our institute, including 6 cases of ependymoma treated by local irradiation and 6 cases of germinoma treated by irradiation of all four cerebral ventricles. IMRT and 3D-CRT treatment plans were made for these 12 cases, with optimization using the same planning conditions as those for PBT. Model cases were also compared using sphere targets with different diameters or locations in the brain, and the normal brain doses with PBT, IMRT and 3D-CRT were compared using the same planning conditions. PBT significantly reduced the average dose to normal brain tissue compared to 3D-CRT and IMRT in all cases. There was no difference between 3D-CRT and IMRT. The average normal brain doses for PBT, 3D-CRT, and IMRT were 5.1–34.8% (median 14.9%), 11.0–48.5% (23.8%), and 11.5–53.1% (23.5%), respectively, in ependymoma cases; and 42.3–61.2% (48.9%), 54.5–74.0% (62.8%), and 56.3–72.1% (61.2%), respectively, in germinoma cases. In the model cases, PBT significantly reduced the average normal brain dose for larger tumors and for tumors located at the periphery of the brain. PBT reduces the average dose to normal brain tissue, compared with 3D-CRT and IMRT. The effect is higher for a tumor that is larger or located laterally.
- 22 February 2017
- Published by Springer Nature in Radiation Oncology
- Vol. 12 (1) , 40
Open AccessA comparative study of dose distribution of PBT, 3D-CRT and IMRT for pediatric brain tumors
It was reported that proton beam therapy (PBT) reduced the normal brain dose compared with X-ray therapy for pediatric brain tumors. We considered whether there was not the condition that PBT was more disadvantageous than intensity modulated photon radiotherapy (IMRT) and 3D conventional radiotherapy (3D-CRT) for treatment of pediatric brain tumors about the dose reduction for the normal brain when the tumor location or tumor size were different. The subjects were 12 patients treated with PBT at our institute, including 6 cases of ependymoma treated by local irradiation and 6 cases of germinoma treated by irradiation of all four cerebral ventricles. IMRT and 3D-CRT treatment plans were made for these 12 cases, with optimization using the same planning conditions as those for PBT. Model cases were also compared using sphere targets with different diameters or locations in the brain, and the normal brain doses with PBT, IMRT and 3D-CRT were compared using the same planning conditions. PBT significantly reduced the average dose to normal brain tissue compared to 3D-CRT and IMRT in all cases. There was no difference between 3D-CRT and IMRT. The average normal brain doses for PBT, 3D-CRT, and IMRT were 5.1–34.8% (median 14.9%), 11.0–48.5% (23.8%), and 11.5–53.1% (23.5%), respectively, in ependymoma cases; and 42.3–61.2% (48.9%), 54.5–74.0% (62.8%), and 56.3–72.1% (61.2%), respectively, in germinoma cases. In the model cases, PBT significantly reduced the average normal brain dose for larger tumors and for tumors located at the periphery of the brain. PBT reduces the average dose to normal brain tissue, compared with 3D-CRT and IMRT. The effect is higher for a tumor that is larger or located laterally.
- 28 July 2016
- Published by Springer Nature in Australasian Physics & Engineering Sciences in Medicine
- Vol. 39 (3) , 687-696
Development of simple high-precision two-dimensional dose-distribution measurement method for proton beam therapy using imaging plate and EBT3
Although there are several two-dimensional (2D) dose-distribution measurement methods using proton beam therapy, they all have drawbacks; hence, there is no standard method established worldwide. The purpose of this study was to develop a simple, high-precision 2D distribution measurement method for proton beam therapy that uses an imaging plate and EBT3. First, we expanded the maximum readable dose (saturation dose) in the imaging plate. The method involves (i) the control of the fading phenomenon by an annealing process and (ii) the control of the photostimulated luminescence (PSL) phenomenon using a longpass filter (LPF). In method (i), upon heating at 80 °C, the PSL became 0.485 times the room temperature, and in method (ii), we attenuated the PSL by a factor of 0.245 using an LPF. Thus, by combining methods (i) and (ii), we expanded the saturation dose to 2 Gy. Thus, it was possible to measure the imaging plate and EBT3 in the same dose range. We simultaneously measured the percent depth dose using imaging plate and EBT3. We defined a correction factor to match the measured values—which had a reduced sensitivity because of the linear energy transfer (LET) dependence of the imaging plate and EBT3—with reference data and developed a correction factor function. Subsequently, by defining the relative LET dependence of imaging plate and EBT3 as the relative sensitivity and converting the relationship imaging plate between the relative sensitivity and correction factor into a function, we obtained a sensitivity-correction function. By employing this function, measurements with the same accuracy as the reference data were performed using the imaging plate and EBT3.
- 28 July 2016
- Published by Springer Nature in Australasian Physics & Engineering Sciences in Medicine
- Vol. 39 (3) , 687-696
Development of simple high-precision two-dimensional dose-distribution measurement method for proton beam therapy using imaging plate and EBT3
Although there are several two-dimensional (2D) dose-distribution measurement methods using proton beam therapy, they all have drawbacks; hence, there is no standard method established worldwide. The purpose of this study was to develop a simple, high-precision 2D distribution measurement method for proton beam therapy that uses an imaging plate and EBT3. First, we expanded the maximum readable dose (saturation dose) in the imaging plate. The method involves (i) the control of the fading phenomenon by an annealing process and (ii) the control of the photostimulated luminescence (PSL) phenomenon using a longpass filter (LPF). In method (i), upon heating at 80 °C, the PSL became 0.485 times the room temperature, and in method (ii), we attenuated the PSL by a factor of 0.245 using an LPF. Thus, by combining methods (i) and (ii), we expanded the saturation dose to 2 Gy. Thus, it was possible to measure the imaging plate and EBT3 in the same dose range. We simultaneously measured the percent depth dose using imaging plate and EBT3. We defined a correction factor to match the measured values—which had a reduced sensitivity because of the linear energy transfer (LET) dependence of the imaging plate and EBT3—with reference data and developed a correction factor function. Subsequently, by defining the relative LET dependence of imaging plate and EBT3 as the relative sensitivity and converting the relationship imaging plate between the relative sensitivity and correction factor into a function, we obtained a sensitivity-correction function. By employing this function, measurements with the same accuracy as the reference data were performed using the imaging plate and EBT3.
- 14 June 2016
- Published by Wiley in Medical Physics
- Vol. 43 (7) , 4150-4162
Dual ring multilayer ionization chamber and theory‐based correction technique for scanning proton therapy
Purpose: To develop a multilayer ionization chamber (MLIC) and a correction technique that suppresses differences between the MLIC and water phantom measurements in order to achieve fast and accurate depth dose measurements in pencil beam scanning proton therapy.Methods: The authors distinguish between a calibration procedure and an additional correction: 1—the calibration for variations in the air gap thickness and the electrometer gains is addressed without involving measurements in water; 2—the correction is addressed to suppress the difference between depth dose profiles in water and in the MLIC materials due to the nuclear interaction cross sections by a semiempirical model tuned by using measurements in water. In the correction technique, raw MLIC data are obtained for each energy layer and integrated after multiplying them by the correction factor because the correction factor depends on incident energy. The MLIC described here has been designed especially for pencil beam scanning proton therapy. This MLIC is called a dual ring multilayer ionization chamber (DRMLIC). The shape of the electrodes allows the DRMLIC to measure both the percentage depth dose (PDD) and integrated depth dose (IDD) because ionization electrons are collected from inner and outer air gaps independently.Results: IDDs for which the beam energies were 71.6, 120.6, 159, 180.6, and 221.4 MeV were measured and compared with water phantom results. Furthermore, the measured PDDs along the central axis of the proton field with a nominal field size of 10 × 10 cm2 were compared. The spread out Bragg peak was 20 cm for fields with a range of 30.6 and 3 cm for fields with a range of 6.9 cm. The IDDs measured with the DRMLIC using the correction technique were consistent with those that of the water phantom; except for the beam energy of 71.6 MeV, all of the points satisfied the 1% dose/1 mm distance to agreement criterion of the gamma index. The 71.6 MeV depth dose profile showed slight differences in the shallow region, but 94.5% of the points satisfied the 1%/1 mm criterion. The 90% ranges, defined at the 90% dose position in distal fall off, were in good agreement with those in the water phantom, and the range differences from the water phantom were less than ±0.3 mm. The PDDs measured with the DRMLIC were also consistent with those that of the water phantom; 97% of the points passed the 1%/1 mm criterion.Conclusions: It was demonstrated that the new correction technique suppresses the difference between the depth dose profiles obtained with the MLIC and those obtained from a water phantom, and a DRMLIC enabling fast measurements of both IDD and PDD was developed. The IDDs and PDDs measured with the DRMLIC and using the correction technique were in good agreement with those that of the water phantom, and it was concluded that the correction technique and DRMLIC are useful for depth dose profile measurements in pencil beam scanning proton therapy.
- 14 June 2016
- Published by Wiley in Medical Physics
- Vol. 43 (7) , 4150-4162
Dual ring multilayer ionization chamber and theory‐based correction technique for scanning proton therapy
Purpose: To develop a multilayer ionization chamber (MLIC) and a correction technique that suppresses differences between the MLIC and water phantom measurements in order to achieve fast and accurate depth dose measurements in pencil beam scanning proton therapy.Methods: The authors distinguish between a calibration procedure and an additional correction: 1—the calibration for variations in the air gap thickness and the electrometer gains is addressed without involving measurements in water; 2—the correction is addressed to suppress the difference between depth dose profiles in water and in the MLIC materials due to the nuclear interaction cross sections by a semiempirical model tuned by using measurements in water. In the correction technique, raw MLIC data are obtained for each energy layer and integrated after multiplying them by the correction factor because the correction factor depends on incident energy. The MLIC described here has been designed especially for pencil beam scanning proton therapy. This MLIC is called a dual ring multilayer ionization chamber (DRMLIC). The shape of the electrodes allows the DRMLIC to measure both the percentage depth dose (PDD) and integrated depth dose (IDD) because ionization electrons are collected from inner and outer air gaps independently.Results: IDDs for which the beam energies were 71.6, 120.6, 159, 180.6, and 221.4 MeV were measured and compared with water phantom results. Furthermore, the measured PDDs along the central axis of the proton field with a nominal field size of 10 × 10 cm2 were compared. The spread out Bragg peak was 20 cm for fields with a range of 30.6 and 3 cm for fields with a range of 6.9 cm. The IDDs measured with the DRMLIC using the correction technique were consistent with those that of the water phantom; except for the beam energy of 71.6 MeV, all of the points satisfied the 1% dose/1 mm distance to agreement criterion of the gamma index. The 71.6 MeV depth dose profile showed slight differences in the shallow region, but 94.5% of the points satisfied the 1%/1 mm criterion. The 90% ranges, defined at the 90% dose position in distal fall off, were in good agreement with those in the water phantom, and the range differences from the water phantom were less than ±0.3 mm. The PDDs measured with the DRMLIC were also consistent with those that of the water phantom; 97% of the points passed the 1%/1 mm criterion.Conclusions: It was demonstrated that the new correction technique suppresses the difference between the depth dose profiles obtained with the MLIC and those obtained from a water phantom, and a DRMLIC enabling fast measurements of both IDD and PDD was developed. The IDDs and PDDs measured with the DRMLIC and using the correction technique were in good agreement with those that of the water phantom, and it was concluded that the correction technique and DRMLIC are useful for depth dose profile measurements in pencil beam scanning proton therapy.