RESUMEN
Electrometers are important devices that are part of the standard dosimetry system. Therefore, we evaluated the variation of electrometer calibration coefficients (kelec) over 1 year in this study. We investigated two types of electrometers: a rate mode and an integrate mode. Each electrometer was connected to a charge generator, a constant charge was applied, and kelec was determined by measuring the current. The current measurements were repeated once a month. For electrometers with multiple ranges, measurements were taken at low and medium ranges. Almost all kelec measurements agreed within 0.2% of the initial measurements. However, the low range of the electrometer with an integrate mode showed seasonal variation, with a variation greater than 0.2%. This study shows that electrometers may exhibit errors that cannot be detected through annual inspections. The importance of quality assurance using a charge generator at one's own institution was demonstrated.
Asunto(s)
Radiometría , Calibración , Radiometría/instrumentación , Radioterapia/instrumentación , Estaciones del AñoRESUMEN
To propose a style transfer model for multi-contrast magnetic resonance imaging (MRI) images with a cycle-consistent generative adversarial network (CycleGAN) and evaluate the image quality and prognosis prediction performance for glioblastoma (GBM) patients from the extracted radiomics features. Style transfer models of T1 weighted MRI image (T1w) to T2 weighted MRI image (T2w) and T2w to T1w with CycleGAN were constructed using the BraTS dataset. The style transfer model was validated with the Cancer Genome Atlas Glioblastoma Multiforme (TCGA-GBM) dataset. Moreover, imaging features were extracted from real and synthesized images. These features were transformed to rad-scores by the least absolute shrinkage and selection operator (LASSO)-Cox regression. The prognosis performance was estimated by the Kaplan-Meier method. For the accuracy of the image quality of the real and synthesized MRI images, the MI, RMSE, PSNR, and SSIM were 0.991 ± 2.10 × 10 - 4 , 2.79 ± 0.16, 40.16 ± 0.38, and 0.995 ± 2.11 × 10 - 4 , for T2w, and .992 ± 2.63 × 10 - 4 , 2.49 ± 6.89 × 10 - 2 , 40.51 ± 0.22, and 0.993 ± 3.40 × 10 - 4 for T1w, respectively. The survival time had a significant difference between good and poor prognosis groups for both real and synthesized T2w (p < 0.05). However, the survival time had no significant difference between good and poor prognosis groups for both real and synthesized T1w. On the other hand, there was no significant difference between the real and synthesized T2w in both good and poor prognoses. The results of T1w were similar in the point that there was no significant difference between the real and synthesized T1w. It was found that the synthesized image could be used for prognosis prediction. The proposed prognostic model using CycleGAN could reduce the cost and time of image scanning, leading to a promotion to build the patient's outcome prediction with multi-contrast images.
RESUMEN
PURPOSE: Lateral response artifact (LRA) is caused by the interaction between film and flatbed scanner in the direction perpendicular to the scanning direction. This can significantly affect the accuracy of patient-specific quality assurance (QA) in cases involving large irradiation fields. We hypothesized that by utilizing the central area of the flatbed scanner, where the magnitude of LRA is relatively small, the LRA could be mitigated effectively. This study proposes a practical solution using the image-stitching technique to correct LRA for patient-specific QA involving large irradiation fields. METHODS: Gafchromic™ EBT4 film and Epson Expression ES-G11000 flatbed scanner were used in this study. The image-stitching algorithm requires a spot between adjacent images to combine them. The film was scanned at three locations on a flatbed scanner, and these images were combined using the image-stitching technique. The combined film dose was then calculated and compared with the treatment planning system (TPS)-calculated dose using gamma analysis (3%/2 mm). Our proposed LRA correction was applied to several films exposed to 18 × 18 cm2 open fields at doses of 200, 400, and 600 cGy, as well as to four clinical Volumetric Modulated Arc Therapy (VMAT) treatment plans involving large fields. RESULTS: For doses of 200, 400, and 600 cGy, the gamma analysis values with and without LRA corrections were 95.7% versus 67.8%, 95.5% versus 66.2%, and 91.8% versus 35.9%, respectively. For the clinical VMAT treatment plan, the average pass rate ± standard deviation in gamma analysis was 94.1% ± 0.4% with LRA corrections and 72.5% ± 1.5% without LRA corrections. CONCLUSIONS: The effectiveness of our proposed LRA correction using the image-stitching technique was demonstrated to significantly improve the accuracy of patient-specific QA for VMAT treatment plans involving large irradiation fields.
Asunto(s)
Algoritmos , Artefactos , Dosimetría por Película , Procesamiento de Imagen Asistido por Computador , Fantasmas de Imagen , Garantía de la Calidad de Atención de Salud , Dosificación Radioterapéutica , Planificación de la Radioterapia Asistida por Computador , Radioterapia de Intensidad Modulada , Humanos , Dosimetría por Película/métodos , Dosimetría por Película/instrumentación , Planificación de la Radioterapia Asistida por Computador/métodos , Procesamiento de Imagen Asistido por Computador/métodos , Garantía de la Calidad de Atención de Salud/normas , Radioterapia de Intensidad Modulada/métodosRESUMEN
BACKGROUND: Monte Carlo simulation code is commonly used for the dose calculation of boron neutron capture therapy. In the past, dose calculation was performed assuming a homogeneous mass density and elemental composition inside the tissue, regardless of the patient's age or sex. Studies have shown that the mass density varies with patient to patient, particularly for those that have undergone surgery or radiotherapy. A method to convert computed tomography numbers into mass density and elemental weights of tissues has been developed and applied in the dose calculation process using Monte Carlo codes. A recent study has shown the variation in the computed tomography number between different scanners for low- and high-density materials. PURPOSE: The aim of this study is to investigate the effect of the elemental composition inside each calculation voxel on the dose calculation and the application of the stoichiometric CT number calibration method for boron neutron capture therapy planning. METHODS: Monte Carlo simulation package Particle and Heavy Ion Transport code System was used for the dose calculation. Firstly, a homogeneous cubic phantom with the material set to ICRU soft tissue (four component), muscle, fat, and brain was modelled and the NeuCure BNCT system accelerator-based neutron source was used. The central axis depth dose distribution was simulated and compared between the four materials. Secondly, a treatment plan of the brain and the head and neck region was simulated using a dummy patient dataset. Three models were generated; (1) a model where only the fundamental materials were considered (simple model), a model where each voxel was assigned a mass density and elemental weight using (2) the Nakao20 model, and (3) the Schneider00 model. The irradiation conditions were kept the same between the different models (irradiation time and irradiation field size) and the near maximum (D1%) and mean dose to the organs at risk were calculated and compared. RESULTS: A maximum percentage difference of approximately 5% was observed between the different materials for the homogeneous phantom. With the dummy patient plan, a large dose difference in the bone (greater than 12%) and region near the low-density material (mucosal membrane, 7%-11%) was found between the different models. CONCLUSIONS: A stoichiometric CT number calibration method using the newly developed Nakao20 model was applied to BNCT dose calculation. The results indicate the importance of calibrating the CT number to elemental composition for each individual CT scanner for the purpose of BNCT dose calculation along with the consideration of heterogeneity of the material composition inside the defined region of interest.
Asunto(s)
Terapia por Captura de Neutrón de Boro , Método de Montecarlo , Fantasmas de Imagen , Dosificación Radioterapéutica , Planificación de la Radioterapia Asistida por Computador , Tomografía Computarizada por Rayos X , Terapia por Captura de Neutrón de Boro/métodos , Calibración , Humanos , Planificación de la Radioterapia Asistida por Computador/métodos , Dosis de Radiación , Neoplasias de Cabeza y Cuello/radioterapia , Neoplasias de Cabeza y Cuello/diagnóstico por imagenRESUMEN
PURPOSE: We created a phantom and analysis program for the assessment of IGRT positional accuracy. We verified the accuracy of analysis and the practicality of this evaluation method at several facilities. METHOD: End-to-end test was performed using an in-house phantom, and EPID images were acquired after displacement by an arbitrary amount using a micrometer, with after image registration as the reference. The difference between the center of the target and the irradiated field was calculated using our in-house analysis program and commercial software. The end-to-end test was conducted at three facilities, and the IGRT positional accuracy evaluation was verified. RESULT: The maximum difference between the displacement of the target determined from the EPID image and the arbitrary amount of micrometer displacement was 0.24 mm for the in-house analysis program and 0.30 mm for the commercial software. The maximum difference between the center of the target and the irradiation field on EPID images acquired at the three facilities was 0.97 mm. CONCLUSION: The proposed evaluation method using our in-house phantom and analysis program can be used for the assessment of IGRT positional accuracy.
Asunto(s)
Radioterapia Guiada por Imagen , Radioterapia Guiada por Imagen/métodos , Fantasmas de Imagen , Programas InformáticosRESUMEN
BACKGROUND: Inadequate computed tomography (CT) number calibration curves affect dose calculation accuracy. Although CT number calibration curves registered in treatment planning systems (TPSs) should be consistent with human tissues, it is unclear whether adequate CT number calibration is performed because CT number calibration curves have not been assessed for various types of CT number calibration phantoms and TPSs. PURPOSE: The purpose of this study was to investigate CT number calibration curves for mass density (ρ) and relative electron density (ρe ). METHODS: A CT number calibration audit phantom was sent to 24 Japanese photon therapy institutes from the evaluating institute and scanned using their individual clinical CT scan protocols. The CT images of the audit phantom and institute-specific CT number calibration curves were submitted to the evaluating institute for analyzing the calibration curves registered in the TPSs at the participating institutes. The institute-specific CT number calibration curves were created using commercial phantom (Gammex, Gammex Inc., Middleton, WI, USA) or CIRS phantom (Computerized Imaging Reference Systems, Inc., Norfolk, VA, USA)). At the evaluating institute, theoretical CT number calibration curves were created using a stoichiometric CT number calibration method based on the CT image, and the institute-specific CT number calibration curves were compared with the theoretical calibration curve. Differences in ρ and ρe over the multiple points on the curve (Δρm and Δρe,m , respectively) were calculated for each CT number, categorized for each phantom vendor and TPS, and evaluated for three tissue types: lung, soft tissues, and bones. In particular, the CT-ρ calibration curves for Tomotherapy TPSs (ACCURAY, Sunnyvale, CA, USA) were categorized separately from the Gammex CT-ρ calibration curves because the available tissue-equivalent materials (TEMs) were limited by the manufacturer recommendations. In addition, the differences in ρ and ρe for the specific TEMs (ΔρTEM and Δρe,TEM , respectively) were calculated by subtracting the ρ or ρe of the TEMs from the theoretical CT-ρ or CT-ρe calibration curve. RESULTS: The mean ± standard deviation (SD) of Δρm and Δρe,m for the Gammex phantom were -1.1 ± 1.2 g/cm3 and -0.2 ± 1.1, -0.3 ± 0.9 g/cm3 and 0.8 ± 1.3, and -0.9 ± 1.3 g/cm3 and 1.0 ± 1.5 for lung, soft tissues, and bones, respectively. The mean ± SD of Δρm and Δρe,m for the CIRS phantom were 0.3 ± 0.8 g/cm3 and 0.9 ± 0.9, 0.6 ± 0.6 g/cm3 and 1.4 ± 0.8, and 0.2 ± 0.5 g/cm3 and 1.6 ± 0.5 for lung, soft tissues, and bones, respectively. The mean ± SD of Δρm for Tomotherapy TPSs was 2.1 ± 1.4 g/cm3 for soft tissues, which is larger than those for other TPSs. The mean ± SD of Δρe,TEM for the Gammex brain phantom (BRN-SR2) was -1.8 ± 0.4, implying that the tissue equivalency of the BRN-SR2 plug was slightly inferior to that of other plugs. CONCLUSIONS: Latent deviations between human tissues and TEMs were found by comparing the CT number calibration curves of the various institutes.
Asunto(s)
Planificación de la Radioterapia Asistida por Computador , Tomografía Computarizada por Rayos X , Humanos , Calibración , Planificación de la Radioterapia Asistida por Computador/métodos , Tomografía Computarizada por Rayos X/métodos , Cabeza , Huesos , Fantasmas de ImagenRESUMEN
Purpose: We evaluated the dosimetric effect of tumor changes in patients with fractionated brain stereotactic radiation therapy (SRT) on the tumor and normal brain using repeat verification magnetic resonance imaging (MRI) in the middle of the treatment period. Methods and Materials: Fifteen large intracranial metastatic lesions with fractionated SRT were scanned employing standardized planning MRI (MRI-1). Repeat verification MRI (MRI-2) were performed during the middle of the irradiation period. Gross tumor volume (GTV) was defined as the volume of the contrast-enhancing lesion on T1-weighted MRI with gadolinium contrast agent. The doses to the tumor and normal brain were evaluated on the MRI-1 scan. Beam configuration and intensity on the initial volumetric modulated arc therapy plan were used to evaluate the dose to the tumor and the normal brain on MRI-2. We evaluated the effect of D98% (percent dose irradiating 98% of the volume) on the GTV using the plans on the MRI-1 and MRI-2 scans. For the normal brain, the V90%, V80%, and V50% (volume of the normal brain receiving >90%, 80%, and 50% of the prescribed dose, respectively) were investigated. Results: Three (20% of the total) and 4 (26% of the total) tumors exhibited volume shrinkage or enlargement changes of >10%. Five (33% of the total) tumors exhibited volume shrinkage and enlargement changes of <10%. Three tumors (20% of the total) showed no volume changes. D98% of the GTV increased in patients with tumor shrinkage because of dose inhomogeneity and decreased in patients with tumor enlargement, with a coefficient of determination of 0.28. The V90%, V80%, and V50% increase with decreasing tumor volumes and were linearly related to the tumor volume difference with a coefficient of determination values of 0.97, 0.98, and 0.97, respectively. Conclusions: Repeat verification MRI for brain fractionated SRT during the treatment period should be considered to reduce the magnitude of target underdosing or normal brain overdosing.
RESUMEN
BACKGROUND AND PURPOSE: The standard dosimetry system of medical accelerators in radiotherapy consists of an ionization chamber, an electrometer, and cables. Guidance for TG-51 reference dosimetry reported that the electrometer correction factor (Pelec ) should be checked every few years. Therefore, continuous Pelec measurements have not been reported. The purpose of this study is to measure the Pelec with a charge generator at our institution and to evaluate variations over time. The measurements are compared with calibration data given by an Accredited Dosimetry Calibration Laboratory (ADCL). MATERIALS AND METHODS: We used four reference-class electrometers: RT521R (RTQM system/EMF Japan), Model 35040 (FLUKE), RAMTEC Duo (Toyo medic), and UNIDOS-E (PTW). Each electrometer was connected to the charge generator, and the required charge was applied. The measurement points used were the same as those used for calibration by the ADCL. From the measured charges at each point, the Pelec was obtained from the slope of the linear regression function. The measurements were repeated over a 3-month period to evaluate variations over time for each electrometer. Additionally, error budgets for the Pelec measurements were estimated, and the overall uncertainty was determined. RESULTS: The measured Pelec values were 1.0000, 0.9995, 1.0009/0.9999, and 0.9995/0.9998 for RT521R, Model 35040, the low/medium (L/M) ranges of RAMTEC Duo, and the L/M ranges of UNIDOS-E, respectively. The measured Pelec values agreed within 0.1% with those given by the ADCL. We found a small drift in the measurements for one electrometer. Additionally, the uncertainty considered was 0.26% for k = 2 (k, coverage factor). CONCLUSION: In this study, stable Pelec values were obtained for four electrometers using a charge generator over a three-month period. The measured Pelec values were within the overall uncertainty stated in the electrometer guidelines. However, performing periodic measurements for the Pelec was able to help in detecting small errors.
Asunto(s)
Radiometría , Humanos , Radiometría/métodos , Calibración , JapónRESUMEN
BACKGROUND/AIM: We measured interfractional liver positional motion in liver stereotactic body radiotherapy (SBRT) with exhaled breath holding (BH) based on kilovoltage (kV) cone-beam computed tomography (CBCT) images. PATIENTS AND METHODS: We collected 528 pre-treatment kV-CBCT images from 132 patients who underwent liver SBRT under exhaled BH using the Abches system, a non-electronic contact-based respiratory monitoring device, and analyzed them to investigate interfractional liver positional motion. Planning computed tomography (CT) scans were obtained using the Abches system when the patients were under exhaled BH. Translational 3-degree-of-freedom (DOF) soft-tissue-based image registration was performed using the kV-CBCT images under exhaled BH after 6-DOF vertebral bone image registration. Interfractional liver positional motions in the left-right (LR), anteroposterior (AP), and craniocaudal (CC) directions were defined based on the differences in the position of the liver relative to the vertebral bones. RESULTS: For all fractions, the absolute mean±standard deviation for the interfractional liver positional motion in the LR, AP, and CC directions was 0.7±1.0 mm, 1.0±1.5 mm, and 2.8±3.1 mm, respectively. The liver interfractional systematic/random positional motions in the LR, AP, and CC directions were 0.9/1.2 mm, 1.4/1.8 mm, and 2.9/3.9 mm, respectively. For all fractions, 100.0%, 98.3%, and 86.9% of the interfractional liver positional motions in the LR, AP, and CC directions, respectively, were less than 5 mm. CONCLUSION: CBCT-guided online correction should be used to correct interfractional liver positions errors present in liver SBRT with exhaled BH.
Asunto(s)
Contencion de la Respiración , Radiocirugia , Humanos , Hígado/diagnóstico por imagen , Movimiento (Física) , Abdomen , Tomografía Computarizada de Haz Cónico/métodos , Radiocirugia/métodosRESUMEN
The purpose of this study was to investigate the impact of scanning orientation and lateral response artifact (LRA) effects on the dose-response of EBT4 films and compare it with that of EBT3 films. Dose-response curves for EBT3 and EBT4 films in red-green-blue (RGB) color channels in portrait orientation were created for unexposed films and for films exposed to doses ranging from 0 to 1 000 cGy. Portrait and landscape orientations of the EBT3 and EBT4 films were scanned to investigate the scanning orientation effect in the red channel. EBT3 and EBT4 films were irradiated to assess the LRA in the red channel using a field size of 15 × 15 cm2 and delivered doses of 200, 400, and 600 cGy. Films were scanned at the edge of the scanner bed, and the measured doses were compared with the treatment planning system (TPS) calculated doses at a position 100 mm lateral to the scanner center. At a dose of 200 cGy, the differences in optical density (OD) in the red, green, and blue color channels between EBT3 and EBT4 films were 0.035 (24.8%), 0.042 (49.7%), and 0.022 (64.4%), respectively. The EBT4 film slightly improved the scanning orientation compared to the EBT3 film. The OD difference in the different scanning orientations for the EBT3 and EBT4 films was 0.015 (6.8%) and 0.007 (3.9%), respectively, at a dose of 200 cGy. This is equivalent to a 20 or 10 cGy variation at a dose of 200 cGy. Compared with the TPS calculation, the measurement doses for EBT3 and EBT4 films irradiated at 200 cGy were approximately 16% and 13% higher, respectively, at the 100 mm off-centered position. The EBT4 film showed an improvement concerning the impact of LRA compared with the EBT3 film. This study demonstrated that the response of EBT4 film to a dose in the blue channel was less sensitive and showed an improvement in the scanning orientation and LRA effects.
Asunto(s)
Artefactos , Dosimetría por Película , Humanos , CalibraciónRESUMEN
BACKGROUND: Artificial intelligence (AI)-based gamma passing rate (GPR) prediction has been proposed as a time-efficient virtual patient-specific QA method for the delivery of volumetric modulation arc therapy (VMAT). However, there is a limitation that the GPR value loses the locational information of dose accuracy. PURPOSE: The objective was to predict the failing points in the gamma distribution and the GPR using a synthesized gamma distribution of VMAT QA with a deep convolutional generative adversarial network (GAN). METHODS: The fluence maps of 270 VMAT beams for prostate cancer were measured using an electronic portal imaging device and analyzed using gamma evaluation with 3%/2-mm, 2%/1-mm, 1%/1-mm, and 1%/0.5-mm tolerances. The 270 gamma distributions were divided into two datasets: 240 training datasets for creating a model and 30 test datasets for evaluation. The image prediction network for the fluence maps calculated by the treatment planning system (TPS) to the gamma distributions was created using a GAN. The sensitivity, specificity, and accuracy of detecting failing points were evaluated using measured and synthesized gamma distributions. In addition, the difference between measured GPR (mGPR) and predicted GPR (pGPR) values calculated from the synthesized gamma distributions was evaluated. RESULTS: The root mean squared errors between mGPR and pGPR were 1.0%, 2.1%, 3.5%, and 3.6% for the 3%/2-mm, 2%/1-mm, 1%/1-mm, and 1%/0.5-mm tolerances, respectively. The accuracies for detecting failing points were 98.9%, 96.9%, 94.7%, and 93.7% for 3%/2-mm, 2%/1-mm, 1%/1-mm, and 1%/0.5-mm tolerances, respectively. The sensitivity and specificity were the highest for 1%/0.5-mm and 3%/2-mm tolerances, which were 82.7% and 99.6%, respectively. CONCLUSIONS: We developed a novel system using a GAN to generate a synthesized gamma distribution-based patient-specific VMAT QA. The system is promising from the point of view of quality assurance in radiotherapy because it shows high performance and can detect failing points.
Asunto(s)
Neoplasias de la Próstata , Radioterapia de Intensidad Modulada , Masculino , Humanos , Radioterapia de Intensidad Modulada/métodos , Inteligencia Artificial , Planificación de la Radioterapia Asistida por Computador/métodos , Dosificación Radioterapéutica , Neoplasias de la Próstata/diagnóstico por imagen , Neoplasias de la Próstata/radioterapia , Garantía de la Calidad de Atención de SaludRESUMEN
This study aims to evaluate the effect of different air computed tomography (CT) numbers of the image value density table (IVDT) on the retrospective dose calculation of head-and-neck (HN) radiotherapy using TomoTherapy megavoltage CT (MVCT) images. The CT numbers of the inside and outside air and each tissue-equivalent plug of the "Cheese" phantom were obtained from TomoTherapy MVCT. Two IVDTs with different air CT numbers were created and applied to MVCT images of the HN anthropomorphic phantom and recalculated by Planned Adaptive to verify dose distribution. We defined the recalculation dose with MVCT images using both inside and outside air of the IVDT as IVDT MVCT inair ${\mathrm{IVDT}}_{\mathrm{MVCT}}^{\mathrm{inair}}$ and IVDT MVCT outair ${\mathrm{IVDT}}_{\mathrm{MVCT}}^{\mathrm{outair}}$ , respectively. Treatment planning doses calculated on kVCT images were compared with those calculated on MVCT images using two different IVDT tables, namely, IVDT MVCT inair ${\mathrm{IVDT}}_{\mathrm{MVCT}}^{\mathrm{inair}}$ and IVDT MVCT outair ${\mathrm{IVDT}}_{\mathrm{MVCT}}^{\mathrm{outair}}$ . The difference between average MVCT numbers ±1 standard deviation on inside and outside air of the calibration phantom was 65 ± 36 HU. This difference in MVCT number of air exceeded the recommendation lung tolerance for dose calculation error of 2%. The dose differences between the planning target volume (PTV): D98% , D50% , D2% and the organ at risk (OAR): Dmax , Dmean recalculated by IVDT MVCT inair ${\mathrm{IVDT}}_{\mathrm{MVCT}}^{\mathrm{inair}}$ and IVDT MVCT outair ${\mathrm{IVDT}}_{\mathrm{MVCT}}^{\mathrm{outair}}$ using MVCT images were a maximum of 0.7% and 1.2%, respectively. Recalculated doses to the PTV and OAR with MVCT showed that IVDT MVCT outair ${\mathrm{IVDT}}_{\mathrm{MVCT}}^{\mathrm{outair}}$ was 0.5%-0.7% closer to the kVCT treatment planning dose than IVDT MVCT inair ${\mathrm{IVDT}}_{\mathrm{MVCT}}^{\mathrm{inair}}$ . This study showed that IVDT MVCT outair ${\mathrm{IVDT}}_{\mathrm{MVCT}}^{\mathrm{outair}}$ was more accurate than IVDT MVCT inair ${\mathrm{IVDT}}_{\mathrm{MVCT}}^{\mathrm{inair}}$ in recalculating the dose HN cases of MVCT using TomoTherapy.
Asunto(s)
Planificación de la Radioterapia Asistida por Computador , Radioterapia de Intensidad Modulada , Humanos , Estudios Retrospectivos , Planificación de la Radioterapia Asistida por Computador/métodos , Tomografía Computarizada por Rayos X/métodos , Tomografía Computarizada de Haz CónicoRESUMEN
Background: The effective atomic numbers obtained from dual-energy computed tomography (DECT) can aid in characterization of materials. In this study, an effective atomic number image reconstructed from a DECT image was synthesized using an equivalent single-energy CT image with a deep convolutional neural network (CNN)-based generative adversarial network (GAN). Materials and methods: The image synthesis framework to obtain the effective atomic number images from a single-energy CT image at 120 kVp using a CNN-based GAN was developed. The evaluation metrics were the mean absolute error (MAE), relative root mean square error (RMSE), relative mean square error (MSE), structural similarity index (SSIM), peak signal-to-noise ratio (PSNR), and mutual information (MI). Results: The difference between the reference and synthetic effective atomic numbers was within 9.7% in all regions of interest. The averages of MAE, RMSE, MSE, SSIM, PSNR, and MI of the reference and synthesized images in the test data were 0.09, 0.045, 0.0, 0.89, 54.97, and 1.03, respectively. Conclusions: In this study, an image synthesis framework using single-energy CT images was constructed to obtain atomic number images scanned by DECT. This image synthesis framework can aid in material decomposition without extra scans in DECT.
RESUMEN
To predict the gamma passing rate (GPR) of the three-dimensional (3D) detector array-based volumetric modulated arc therapy (VMAT) quality assurance (QA) for prostate cancer using a convolutional neural network (CNN) with the 3D dose distribution. One hundred thirty-five VMAT plans for prostate cancer were selected: 110 plans were used for training and validation, and 25 plans were used for testing. Verification plans were measured using a helical 3D diode array (ArcCHECK). The dose distribution on the detector element plane of these verification plans was used as input data for the CNN model. The measured GPR (mGPR) values were used as the training data. The CNN model comprises eighteen layers and predicted GPR (pGPR) values. The mGPR and pGPR values were compared, and a cumulative frequency histogram of the prediction error was created to clarify the prediction error tendency. The correlation coefficients of pGPR and mGPR were 0.67, 0.69, 0.66, and 0.73 for 3%/3-mm, 3%/2-mm, 2%/3-mm, and 2%/2-mm gamma criteria, respectively. The respective mean±standard deviations of pGPR-mGPR were -0.87±2.18%, -0.65±2.93%, -0.44±2.53%, and -0.71±3.33%. The probabilities of false positive error cases (pGPR < mGPR) were 72%, 60%, 68%, and 56% for each gamma criterion. We developed a deep learning-based prediction model of the 3D detector array-based VMAT QA for prostate cancer, and evaluated the accuracy and tendency of prediction GPR. This model can provide a proactive estimation for the results of the patient-specific QA before the verification measurement.
Asunto(s)
Aprendizaje Profundo , Neoplasias de la Próstata , Radioterapia de Intensidad Modulada , Masculino , Humanos , Radioterapia de Intensidad Modulada/métodos , Planificación de la Radioterapia Asistida por Computador/métodos , Garantía de la Calidad de Atención de Salud , Neoplasias de la Próstata/diagnóstico por imagen , Neoplasias de la Próstata/radioterapiaRESUMEN
A combination of a three-dimensional conformal radiation therapy (3D-CRT) plan with a dose gradient of the chest wall area and a volumetric modulated arc therapy (VMAT) plan of the supraclavicular area might improve the dose distribution robustness in the junction. To investigate the impact of patient motion on the dose distribution, hybrid 3D-CRT and VMAT plans were recalculated by shifting the isocenter of the VMAT plan. Compared to the nominal plan, the target D98% for high- vs low-dose gradients decreased by 24% vs 12%. Hybrid VMAT with a low-dose gradient 3D-CRT plan was found to be robust towards patient motion.
RESUMEN
PURPOSE: Treatment planning for ion therapy involves the conversion of computed tomography number (CTN) into a stopping-power ratio (SPR) relative to water. The purpose of this study was to create a CTN-to-SPR calibration table using a stoichiometric CTN calibration model with a three-parameter fit model for ion therapy, and to demonstrate its effectiveness by comparing it with a conventional stoichiometric CTN calibration model. METHODS: We inserted eight tissue-equivalent materials into a CTN calibration phantom and used six CT scanners at five radiotherapy institutes to scan the phantom. We compared the theoretical CTN-to-SPR calibration tables created using the three-parameter fit and conventional models to the measured CTN-to-SPR calibration table in three tissue types: lung, adipose/muscle, and cartilage/spongy bone. We validated the estimated SPR differences in all cases and in a worst-case scenario, which revealed the largest estimated SPR difference in lung tissue. RESULTS: For all cases, the means ± standard deviations of the estimated SPR difference for the three-parameter fit method model were -0.1 ± 1.0%, 0.3 ± 0.7%, and 2.4 ± 0.6% for the lung, adipose/muscle, and cartilage/spongy bone, respectively. For the worst-case scenario, the estimated SPR differences of the conventional and the three-parameter fit models were 2.9% and -1.4% for the lung tissue, respectively. CONCLUSIONS: The CTN-to-SPR calibration table of the three-parameter fit model was consistent with that of the measurement and decreased the calibration error for low-density tissues, even for the worst-case scenario.
Asunto(s)
Tomografía Computarizada por Rayos X , Agua , Calibración , Fantasmas de Imagen , Planificación de la Radioterapia Asistida por Computador/métodos , Tomografía Computarizada por Rayos X/métodosRESUMEN
PURPOSE: Adaptive radiotherapy requires auto-segmentation in patients with head and neck (HN) cancer. In the current study, we propose an auto-segmentation model using a generative adversarial network (GAN) on magnetic resonance (MR) images of HN cancer for MR-guided radiotherapy (MRgRT). MATERIAL AND METHODS: In the current study, we used a dataset from the American Association of Physicists in Medicine MRI Auto-Contouring (RT-MAC) Grand Challenge 2019. Specifically, eight structures in the MR images of HN region, namely submandibular glands, lymph node level II and level III, and parotid glands, were segmented with the deep learning models using a GAN and a fully convolutional network with a U-net. These images were compared with the clinically used atlas-based segmentation. RESULTS: The mean Dice similarity coefficient (DSC) of the U-net and GAN models was significantly higher than that of the atlas-based method for all the structures (p < 0.05). Specifically, the maximum Hausdorff distance (HD) was significantly lower than that in the atlas method (p < 0.05). Comparing the 2.5D and 3D U-nets, the 3D U-net was superior in segmenting the organs at risk (OAR) for HN patients. The DSC was highest for 0.75-0.85, and the HD was lowest within 5.4 mm of the 2.5D GAN model in all the OARs. CONCLUSIONS: In the current study, we investigated the auto-segmentation of the OAR for HN patients using U-net and GAN models on MR images. Our proposed model is potentially valuable for improving the efficiency of HN RT treatment planning.
Asunto(s)
Aprendizaje Profundo , Neoplasias de Cabeza y Cuello , Neoplasias de Cabeza y Cuello/diagnóstico por imagen , Neoplasias de Cabeza y Cuello/radioterapia , Humanos , Procesamiento de Imagen Asistido por Computador/métodos , Imagen por Resonancia Magnética , Órganos en RiesgoRESUMEN
OBJECTIVE: The current study aims to propose the auto-segmentation model on CT images of head and neck cancer using a stepwise deep neural network (stepwise-net). MATERIAL AND METHODS: Six normal tissue structures in the head and neck region of 3D CT images: Brainstem, optic nerve, parotid glands (left and right), and submandibular glands (left and right) were segmented with deep learning. In addition to a conventional convolutional neural network (CNN) on U-net, a stepwise neural network (stepwise-network) was developed. The stepwise-network was based on 3D FCN. We designed two networks in the stepwise-network. One is identifying the target region for the segmentation with the low-resolution images. Then, the target region is cropped, which used for the input image for the prediction of the segmentation. These were compared with a clinical used atlas-based segmentation. RESULTS: The DSCs of the stepwise-net was significantly higher than the atlas-based method for all organ at risk structures. Similarly, the JSCs of the stepwise-net was significantly higher than the atlas-based methods for all organ at risk structures. The Hausdorff distance (HD) was significantly smaller than the atlas-based method for all organ at-risk structures. For the comparison of the stepwise-net and U-net, the stepwise-net had a higher DSC and JSC and a smaller HD than the conventional U-net. CONCLUSIONS: We found that the stepwise-network plays a role is superior to conventional U-net-based and atlas-based segmentation. Our proposed model that is a potentially valuable method for improving the efficiency of head and neck radiotherapy treatment planning.
RESUMEN
The purpose of this study was to extract the three-dimensional (3D) vector of the baseline drift baseline drift vector (BDV) of the specific points on the body surface and to demonstrate the importance of the 3D tracking of the body surface. Our system consisted of a near-infrared camera (NIC: Kinect V2) and software that recognized and tracked blue stickers as markers. We acquired 3D coordinates of 30 markers stuck on the body surface for 30 min for eight healthy volunteers and developed a simple technique to extract the BDV. The BDV on the sternum, rib, and abdomen was extracted from the measured data. BDV per min. was analyzed to estimate the frequency to exceed a given tolerance. Also, the correlation among BDVs for multiple body sites was analyzed. The longitudinal baseline drift was observed in the BDV of healthy volunteers. Among the eight volunteers, the maximum probability that the BDV per min. exceeded the tolerance of 1 mm and 2 mm was 30% and 15%, respectively. The correlation among BDVs of multiple body sites suggested a potential feasibility to distinguish the translational movement of the whole area and the respiratory movement. In conclusion, we constructed the 3D tracking system of multiple points on the body surface using a noninvasive NIC at a low cost and established the method to extract the BDV. The existence of the longitudinal baseline drift showed the importance of the 3D tracking in the body surface.