OpenKBP-Opt: an international and reproducible evaluation of 76 knowledge-based planning pipelines.

Objective.To establish an open framework for developing plan optimization models for knowledge-based planning (KBP).Approach.Our framework includes radiotherapy treatment data (i.e. reference plans) for 100 patients with head-and-neck cancer who were treated with intensity-modulated radiotherapy. That data also includes high-quality dose predictions from 19 KBP models that were developed by different research groups using out-of-sample data during the OpenKBP Grand Challenge. The dose predictions were input to four fluence-based dose mimicking models to form 76 unique KBP pipelines that generated 7600 plans (76 pipelines × 100 patients). The predictions and KBP-generated plans were compared to the reference plans via: the dose score, which is the average mean absolute voxel-by-voxel difference in dose; the deviation in dose-volume histogram (DVH) points; and the frequency of clinical planning criteria satisfaction. We also performed a theoretical investigation to justify our dose mimicking models.Main results.The range in rank order correlation of the dose score between predictions and their KBP pipelines was 0.50-0.62, which indicates that the quality of the predictions was generally positively correlated with the quality of the plans. Additionally, compared to the input predictions, the KBP-generated plans performed significantly better (P< 0.05; one-sided Wilcoxon test) on 18 of 23 DVH points. Similarly, each optimization model generated plans that satisfied a higher percentage of criteria than the reference plans, which satisfied 3.5% more criteria than the set of all dose predictions. Lastly, our theoretical investigation demonstrated that the dose mimicking models generated plans that are also optimal for an inverse planning model.Significance.This was the largest international effort to date for evaluating the combination of KBP prediction and optimization models. We found that the best performing models significantly outperformed the reference dose and dose predictions. In the interest of reproducibility, our data and code is freely available.

Stoichiometric CT number calibration using three-parameter fit model for ion therapy.

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.

Evaluation of two independent dose prediction methods to personalize the automated radiotherapy planning process for prostate cancer.

BACKGROUND AND PURPOSE: Currently, automatic approaches for radiotherapy planning are widely used, however creation of high quality treatment plans is still challenging. In this study, two independent dose prediction methods were used to personalize the initial settings for the automated planning template for optimizing prostate cancer treatment plans. This study evaluated the dose metrics of these plans comparing both methods with the current clinical automated prostate cancer treatment plans. MATERIAL AND METHODS: Datasets of 20 high-risk prostate cancer treatment plans were taken from our clinical database. The prescription dose for these plans was 70 Gy given in fractions of 2.5 Gy. Plans were replanned using the current clinical automated treatment and compared with two personalized automated planning methods. The feasibility dose volume histogram (FDVH) and modified filter back projection (mFBP) methods were used to calculate independent dose predictions. Parameters for the initial objective values of the planning template were extracted from these predictions and used to personalize the optimization of the automated planning process. RESULTS: The current automated replanned clinical plans and the automated plans optimized with the personalized template methods fulfilled the clinical dose criteria. For both methods a reduction in the average mean dose of the rectal wall was found, from 22.5 to 20.1 Gy for the FDVH and from 22.5 to 19.6 Gy for the mFBP method. CONCLUSIONS: With both dose-prediction methods the initial settings of the template could be personalized. Hereby, the average dose to the rectal wall was reduced compared to the standard template method.

Verification system for intensity-modulated radiation therapy with scintillator.

In the preparation of intensity-modulated radiation therapy (IMRT), patient-specific verification is widely employed to optimize the treatment. To accurately estimate the accumulated dose and obtain the field-by-field or segment-by-segment verification, an original IMRT verification tool using scintillator light and an analysis workflow was developed in this study. The raw light distribution was calibrated with respect to the irradiated field size dependency and light diffusion in the water. The calibrated distribution was converted to dose quantity and subsequently compared with the results of the clinically employed plan. A criterion of 2-mm dose-to-agreement and 3% dose difference was specified in the gamma analysis with a 10% dose threshold. By applying the light diffusion calibration, the maximum dose difference was corrected from 7.7 cGy to 3.9 cGy around the field edge for a 60 cGy dose, 7 × 7 cm2 irradiation field, and 10 MV beam energy. Equivalent performance was confirmed in the chromodynamic film. The average dose difference and gamma pass rate of the accumulated dose distributions in six patients were 0.8 ± 4.5 cGy and 97.4%, respectively. In the field-by-field analysis, the average dose difference and gamma pass rate in seven fields of Patient 1 were 0.2 ± 1.2 cGy and 93.9%, respectively. In the segment-by-segment analysis, the average dose difference and gamma pass rate in nine segments of Patient 1 and a 305° gantry angle were - 0.03 ± 0.2 cGy and 93.9%, respectively. This system allowed the simultaneous and independent analysis of each field or segment in the accumulated dose analysis.

PD-L1 expression in tongue squamous cell carcinoma.

PURPOSE: Immune checkpoint proteins programmed cell death 1 (PD-1) and programmed cell death ligand 1 (PD-L1) are important therapeutic targets for head and neck cancer. This large-scale case study aimed to analyze tongue squamous cell carcinomas (SCCs) and evaluate the correlation between PD-L1 expression and clinical prognosis. So far, this study is the largest case study on PD-L1 expression in tongue SCCs. METHODS: This is a case-control study that analyzed 121 tongue SCCs. Paraffin-embedded sections and clinical data were obtained retrospectively and immunohistochemistry with PD-L1 was performed. RESULTS: 11.6% contained ≥ 50% of PD-L1-positive cells, 57.1% of these cases had a poor prognosis with nodal metastasis. Among cases of T1/2 primary lesions with nodal metastasis, cases of high PD-L1 expression had a significantly shorter disease-free survival than cases of no PD-L1 expression (p = 0.018). The hazard ratio for high PD-L1 expression was 3.21 (95 per cent CI, 1.26-8.72) compared with no PD-L1 expression after adjusting for other factors. CONCLUSIONS: These data indicate that PD-L1 upregulation in tongue SCCs is associated with a more advanced stage and shorter disease-free survival. PD-1/PD-L1 inhibitors might hence constitute potential adjuvant therapy for tongue SCCs with PD-L1 upregulation.

Evaluation of optimization workflow using custom-made planning through predicted dose distribution for head and neck tumor treatment.

PURPOSE: Lack of a reference dose distribution is one of the challenges in the treatment planning used in volumetric modulated arc therapy because numerous manual processes result from variations in the location and size of a tumor in different cases. In this study, a predicted dose distribution was generated using two independent methods. Treatment planning using the predicted distribution was compared with the clinical value, and its efficacy was evaluated. METHODS: Computed tomography scans of 81 patients with oropharynx or hypopharynx tumors were acquired retrospectively. The predicted dose distributions were determined using a modified filtered back projection (mFBP) and a hierarchically densely connected U-net (HD-Unet). Optimization parameters were extracted from the predicted distribution, and the optimized dose distribution was obtained using a commercial treatment planning system. RESULTS: In the test data from ten patients, significant differences between the mFBP and clinical plan were observed for the maximum dose of the brain stem, spinal cord, and mean dose of the larynx. A significant difference between the dose distributions from the HD-Unet dose and the clinical plan was observed for the mean dose of the left parotid gland. In both cases, the equivalent coverage and flatness of the clinical plan were observed for the tumor target. CONCLUSIONS: The predicted dose distribution was generated using two approaches. In the case of the mFBP approach, no prior learning, such as deep learning, is required; therefore, the accuracy and efficiency of treatment planning will be improved even for sites where sufficient training data are unavailable.

A Single Institution's Experience of Definitive Radiotherapy Using Volumetric-modulated Arc Therapy for Hypopharyngeal Cancers.

BACKGROUND/AIM: The present study aimed to analyze the treatment outcome after definitive radiotherapy (dRT) using volumetric-modulated arc therapy (VMAT) in patients with hypopharyngeal cancer (HPC), including an examination of late toxicities. PATIENTS AND METHODS: A total of 62 patients with HPC, who underwent dRT using VMAT, were analyzed. Overall survival (OS), progression-free survival (PFS), laryngoesophageal dysfunction-free survival (LEDFS), and locoregional control (LRC) were calculated. RESULTS: The median follow-up period was 49 months. The 3- and 5-year OS, PFS, LEDFS, and LRC rates were 77% and 60%, 61% and 56%, 66% and 53%, and both 79%, respectively. Regarding late toxicities, 11 (17.7%) patients developed grade ≥2 late toxicity. Grade 3 dysphagia was observed in 4 (6.5%) patients, and grade 2 xerostomia in 6 (9.7%). CONCLUSION: VMAT was an effective treatment for HPC, with a low incidence of late toxicities.

Evaluation of metal artefact techniques with same contrast scale for different commercially available dual-energy computed tomography scanners.

The aim of our study is to evaluate the metal artefact reduction techniques with the same contrast scale for different vendors' dual-energy CT (DECT): kV-CT image with metal artefact reduction method and monoenergetic CT image using Canon's DECT, and monoenergetic CT image with metal artefact reduction method using GE's DECT. The kV-CT image and DECT scans were performed with the water-based polymethyl methacrylate phantom with various metal materials (brass, aluminium, copper, stainless steel, steel, lead, and titanium). Two types of metal artefact reduction (MAR) algorithm with the monoenergetic CT images were used. Smart MAR implemented by GE and the kV-CT images with MAR algorithms. Single-energy metal artefact reduction (SEMAR), implemented by Canon, was reconstructed. The artefact index was evaluated using the converted electron density values from the kV-CT and DECT images. The artefact index with all material inserts in the monoenergetic CT images were smallest at 70-90 keV for Canon and 140 keV for GE. The artefact index without SEMAR was larger than that with SEMAR for the 80 and 135-kV CT images. In the comparison of the artefact index for the converted electron density images from the 80 and 135-kV CT images with SEMAR, as well as the monoenergetic CT images with and without MAR, the monoenergetic CT image at 140 keV with MAR showed a reduction. In the comparison of the monoenergetic CT images at 140 keV and other energy ranges without and with Smart MAR, there was no statistically significant difference (P < 0.05) for all-metal inserts at more than 100 keV for Canon's DECT and 70 keV for GE's DECT. The metal artefact could be reduced by using a monoenergetic CT image at high energy with MAR algorithm. The metal artefact for the different-contrast-scale images can be compared on the same contrast scale by the electron density conversion method.

Development of a CT number calibration audit phantom in photon radiation therapy: A pilot study.

PURPOSE: In photon radiation therapy, computed tomography (CT) numbers are converted into values for mass density (MD) or relative electron density to water (RED). CT-MD or CT-RED calibration tables are relevant for human body dose calculation in an inhomogeneous medium. CT-MD or CT-RED calibration tables are influenced by patient imaging (CT scanner manufacturer, scanning parameters, and patient size), the calibration process (tissue-equivalent phantom manufacturer, and selection of tissue-equivalent material), differences between tissue-equivalent materials and standard tissues, and the dose calculation algorithm applied; however, a CT number calibration audit has not been established. The purposes of this study were to develop a postal audit phantom, and to establish a CT number calibration audit process. METHODS: A conventional stoichiometric calibration conducts a least square fit of the relationships between the MD, material weight, and measured CT number, using two parameters. In this study, a new stoichiometric CT number calibration scheme has been empirically established, using three parameters to harmonize the calculated CT number with the measured CT number for air and lung tissue. In addition, the suitable material set and the minimal number of materials required for stoichiometric CT number calibration were determined. The MDs and elemental weights from the International Commission on Radiological Protection Publication 110 were used as standard tissue data, to generate the CT-MD and CT-RED calibration tables. A small-sized, CT number calibration phantom was developed for a postal audit, and stoichiometric CT number calibration with the phantom was compared to the CT number calibration tables registered in the radiotherapy treatment planning systems (RTPSs) associated with five radiotherapy institutions. RESULTS: When a least square fit was performed for the stoichiometric CT number calibration with the three parameters, the calculated CT number showed better agreement with the measured CT number. We established stoichiometric CT number calibration using only two materials because the accuracy of the process was determined not by the number of used materials but by the number of elements contained. The stoichiometric CT number calibration was comparable to the tissue-substitute calibration, with a dose difference less than 1%. An outline of the CT number calibration audit was demonstrated through a multi-institutional study. CONCLUSIONS: We established a new stoichiometric CT number calibration method for validating the CT number calibration tables registered in RTPSs. We also developed a CT number calibration phantom for a postal audit, which was verified by the performances of multiple CT scanners located at several institutions. The new stoichiometric CT number calibration has the advantages of being performed using only two materials, and decreasing the difference between the calculated and measured CT numbers for air and lung tissue. In the future, a postal CT number calibration audit might be achievable using a smaller phantom.

MACC1 expression is an indicator of recurrence in early-stage glottic cancer.

BACKGROUND: Metastasis-associated in colon cancer 1 (MACC1) has been reported to be an independent indicator of poor prognoses in some kinds of cancer due to disease metastasis or recurrence. We investigated the correlation between MACC1 expression and the prognosis of glottic cancer. METHODS: Paraffin-embedded, early-stage (I or II) glottic cancer specimens (n = 52) were immunohistochemically analyzed to explore MACC1 expression. The clinical records associated with each case were also examined. Recurrence-free survival (RFS) was estimated using the Kaplan-Meier method, and between-group RFS differences were assessed using the log-rank test. The multivariate analyses were evaluated using the Cox's proportional-hazard model. RESULTS: Patients were treated with only radiotherapy (RT) (n = 37, including 18 with T1 disease and 19 with T2 disease), or with chemoradiotherapy (CRT) (n = 15, including 1 with T1 disease and 14 with T2 disease). Eleven patients demonstrated local recurrence and two patients experienced cervical lymph node recurrence. Tumor specimens were MACC1-positive in 9 of the 13 (69.2%) patients with local or neck recurrence, and 7 of the 11 (63.6%) patients with local recurrence. The RFS rate of patients who were treated with only RT was significantly lower than that of patients who were treated with CRT (P = 0.0243). The RFS rate was significantly lower in cases with MACC1 expression than in those without MACC1 expression (P = 0.0003). Multivariate analysis revealed that MACC1 expression was an independent risk factor of local recurrence (P = 0.0016). CONCLUSION: MACC1 is an independent indicator of recurrence related to RFS in early-stage glottic cancer.

Predictive gamma passing rate for three-dimensional dose verification with finite detector elements via improved dose uncertainty potential accumulation model.

PURPOSE: We aim to develop a method to predict the gamma passing rate (GPR) of a three-dimensional (3D) dose distribution measured by the Delta4 detector system using the dose uncertainty potential (DUP) accumulation model. METHODS: Sixty head-and-neck intensity-modulated radiation therapy (IMRT) treatment plans were created in the XiO treatment planning system. All plans were created using nine step-and-shoot beams of the ONCOR linear accelerator. Verification plans were created and measured by the Delta4 system. The planar DUP (pDUP) manifesting on a field edge was generated from the segmental aperture shape with a Gaussian folding on the beam's-eye view. The DUP at each voxel ( u ) was calculated by projecting the pDUP on the Delta4 phantom with its attenuation considered. The learning model (LM), an average GPR as a function of the DUP, was approximated by an exponential function a GPR u = e - q u to compensate for the low statistics of the learning data due to a finite number of the detectors. The coefficient q was optimized to ensure that the difference between the measured and predicted GPRs ( d GPR ) was minimized. The standard deviation (SD) of the d GPR was evaluated for the optimized LM. RESULTS: It was confirmed that the coefficient q was larger for tighter tolerance. This result corresponds to the expectation that the attenuation of the a GPR u will be large for tighter tolerance. The p GPR and m GPR were observed to be proportional for all tolerances investigated. The SD of d GPR was 2.3, 4.1, and 6.7% for tolerances of 3%/3 mm, 3%/2 mm, 2%/2 mm, respectively. CONCLUSION: The DUP-based predicting method of the GPR was extended to 3D by introducing DUP attenuation and an optimized analytical LM to compensate for the low statistics of the learning data due to a finite number of detector elements. The precision of the predicted GPR is expected to be improved by improving the LM and by involving other metrics.

Tolerance levels of mass density for CT number calibration in photon radiation therapy.

Computed tomography (CT) data are required to calculate the dose distribution in a patient's body. Generally, there are two CT number calibration methods for commercial radiotherapy treatment planning system (RTPS), namely CT number-relative electron density calibration (CT-RED calibration) and CT number-mass density calibration (CT-MD calibration). In a previous study, the tolerance levels of CT-RED calibration were established for each tissue type. The tolerance levels were established when the relative dose error to local dose reached 2%. However, the tolerance levels of CT-MD calibration are not established yet. We established the tolerance levels of CT-MD calibration based on the tolerance levels of CT-RED calibration. In order to convert mass density (MD) to relative electron density (RED), the conversion factors were determined with adult reference computational phantom data available in the International Commission on Radiological Protection publication 110 (ICRP-110). In order to validate the practicability of the conversion factor, the relative dose error and the dose linearity were validated with multiple RTPSes and dose calculation algorithms for two groups, namely, CT-RED calibration and CT-MD calibration. The tolerance levels of CT-MD calibration were determined from the tolerance levels of CT-RED calibration with conversion factors. The converted RED from MD was compared with actual RED calculated from ICRP-110. The conversion error was within ±0.01 for most standard organs. It was assumed that the conversion error was sufficiently small. The relative dose error difference for two groups was less than 0.3% for each tissue type. Therefore, the tolerance levels for CT-MD calibration were determined from the tolerance levels of CT-RED calibration with the conversion factors. The MD tolerance levels for lung, adipose/muscle, and cartilage/spongy-bone corresponded to ±0.044, ±0.022, and ±0.045 g/cm3 , respectively. The tolerance levels were useful in terms of approving the CT-MD calibration table for clinical use.

KRAS mutations in tongue squamous cell carcinoma.

Background: p16INK4a (p16) expression in tongue cancer (TC) is reportedly not associated with human papilloma virus (HPV). Mutations of KRAS in cancer cells are most frequently observed within codon 12. However, few reports have investigated the association between KRAS mutations and p16 status in TC. Objectives: This study aimed to evaluate the influence of KRAS mutations on TC. Methods: Clinical records and surgically resected specimens of 85 TC patients were analyzed. Tumor samples were analyzed for mutations of KRAS located within codons 12 and 13. p16 staining was performed and considered positive in cases with moderate to strong nuclear and cytoplasmic staining. Results: Positive p16 staining was observed in 10 cases (11.8%). A KRAS mutation was detected in one case (1.2%). The case with KRAS mutation showed negative p16 staining. Despite being at an early stage, the patient died of lung metastasis at 43 months from initial treatment. Conclusions and Significance: KRAS mutations are not associated with p16 expression in TC and may predict poor prognosis in TC patients. Further analysis of mutations in regions other than codons 12 and 13 of KRAS will be necessary to determine the relationship between KRAS mutations and prognosis of this disease.

Development of geometrically ideal dose distribution as a reference for treatment planning in VMAT using filtered back-projection method.

PURPOSE: To determine optimal dose distribution in the treatment planning of volumetric modulated arc therapy (VMAT), a virtually ideal dose distribution was developed as a reference by applying filtered back-projection method. METHODS: Delineated structures in patient CT scans were identified using a treatment planning system. The projection of the planning target volume (PTV) was calculated along the X-ray direction for each angle of rotation. Each projection was Fourier transformed to the frequency space; a Shepp-Logan filter was applied, then an inverse Fourier transformation was performed. As the dose irradiation cannot assume a negative value, the filtered projections were shifted using the minimum value inside of the PTV. All values outside of the PTV were set to zero. The corrected filtered projections were then multiplied by the tissue-maximum ratio according to each voxel depth from the surface of the body to simulate X-ray attenuation. Finally, the distributions of multiple rotational angles were convolved to simulate the dose distribution of the VMAT. RESULTS: Ideal dose distributions were generated with sufficient uniformity inside of the PTV. Dose spreading except for the PTV due to external irradiation was reproduced in the case of a brain tumor. A reference dose distribution including OAR sparing was produced. The efficacy of this process as a target for optimum planning was confirmed. CONCLUSION: Using applied filtered back-projection, the ideal dose distribution, which excluded some device-oriented restrictions, was generated. This application will provide support for the determination of VMAT planning quality by providing reference dose distributions.

Young adult patients with squamous cell carcinoma of the tongue strongly express p16 without human papillomavirus infection.

BACKGROUND: Long-term smoking and drinking are known to contribute to the onset of tongue cancer (TC). However, the increasing incidence of TC in younger adults has been suggested to be associated with other factors. OBJECTIVES: The present study investigated the relationship between TC and human papillomavirus (HPV) infection. MATERIAL AND METHODS: Clinical records and surgically resected specimens from 86 patients (<40-years-old, n = 12; ≥40-years-old, n = 74) with TC were analyzed. Strong nuclear and cytoplasmic p16 staining was considered positive. HPV DNA (high-risk subtypes: 16, 18, 31, 33, 35, 52b, and 58; low-risk subtypes: 6 and 11) was detected using consensus primer-mediated polymerase chain reaction. RESULTS: Strong p16 expression was observed in 10 (11.6%) patients. HPV DNA was detected in 9 (10.5%) patients (high-risk subtypes, n = 2; low-risk subtypes, n = 7). Strong p16 expression was observed more frequently among younger adults than among older adults (33.3% vs. 8.1%; p = .045). p16 staining did not correlate with the detection of HPV DNA (correlation coefficient, 0.113; p = .300). CONCLUSIONS AND SIGNIFICANCE: In TC, p16 expression was not associated with HPV infection, suggesting that it may be caused by a different mechanism.

Commissioning of a respiratory gating system involving a pressure sensor in carbon-ion scanning radiotherapy.

This study reports the commissioning methodology and results of a respiratory gating system [AZ - 733 V/733 VI (Anzai Medical Co., Japan)] using a pressure sensor in carbon-ion scanning radiotherapy. Commissioning includes choosing a location and method for pressure sensor installation, delay time measurement of the system, and the final flow test. Additionally, we proposed a methodology for the determination of a threshold level of generating an on/off gate for the beam to the respiratory waveform, which is important for clinical application. Regarding the location and method for installation of the pressure sensor, the actual person's abdomen, back of the body position, and supine/prone positioning were checked. By comparing the motion between the pressure sensor output and the reference LED sensor motion, the chest rear surface was shown to be unsuitable for the sensor installation, due to noise in the signal caused by the cardiac beat. Regarding delay time measurement of the system, measurements were performed for the following four steps: (a). Actual motion to wave signal generation; (b). Wave signal to gate signal generation; (c). Gate signal to beam on/off signal generation; (d). Beam on/off signal to the beam irradiation. The total delay time measured was 46 ms (beam on)/33 ms (beam off); these were within the prescribed tolerance time (<100 ms). Regarding the final flow test, an end-to-end test was performed with a patient verification system using an actual carbon-ion beam; the respiratory gating irradiation was successfully performed, in accordance with the intended timing. Finally, regarding the method for determining the threshold level of the gate generation of the respiration waveform, the target motion obtained from 4D-CT was assumed to be correlated with the waveform obtained from the pressure sensor; it was used to determine the threshold value in amplitude direction.

Computed Tomography Findings for Diagnosing Follicular Thyroid Neoplasms.

Since no diagnostic method has been established to distinguish follicular thyroid carcinoma (FTC) from follicular thyroid adenoma (FTA), surgery has been the only way to reach a diagnosis of follicular neoplasm. Here we investigated the computed tomography (CT) features of follicular neoplasms, toward the goal of being able to identify specific CT features allowing the preoperative differentiation of FTC from FTA. We retrospectively analyzed the cases of 205 patients who underwent preoperative CT of the neck and were histopathologically diagnosed with FTC (n=31) or FTA (n=174) after surgery between January 2002 and June 2016 at several hospitals in Japan. In each of these 205 cases, non-enhanced and contrast-enhanced CT images were obtained, and we analyzed the CT features. On univariate analysis, inhomogeneous features of tumor lesions on contrast-enhanced CT were more frequently observed in FTC than in FTA (p=0.0032). A multivariate analysis identified inhomogeneous features of tumor lesions on contrast-enhanced CT images as an independent variable indicative of FTC (p=0.0023). CT thus offers diagnostic assistance in distinguishing FTC from FTA.

Evaluation of optimization workflow using design of experiment (DoE) for various field configurations in volumetric-modulated arc therapy.

PURPOSE: In volumetric-modulated arc therapy (VMAT), field configurations such as couch or arc angles are defined manually or using a template. A field configuration is reselected through trial-and-error in the case of undesirable resultant planning. To efficiently plan for desirable quality, configurations should be assessed before dose calculation. Design of experiments (DoE) is an optimization technique that efficiently reveals the influence of inputs on outputs. We developed an original tool using DoE to determine the field configuration selection and evaluated the efficacy of this workflow for clinical practice. METHODS: Computed-tomography scans of 17 patients and target structures were acquired retrospectively from a brain tumor treated using a dual-arc VMAT plan. The configurations of the couch, arc, collimator angles, field sizes, and beam energy were determined using DoE. The resultant dose distributions obtained using the DoE-selected configuration were compared with the clinical plan. RESULTS: The averaged differences between the DoE and clinical plan for 17 patients of doses to 50% of the planning target volume (PTV-D50%), Brain-D60%, Brain-D30%, Brain stem-D1%, Left eye-D1%, Right eye-D1%, Optic nerve-D1%, and Chiasm-D1% were 0.2 ±â€¯0.5%, -1.0 ±â€¯4.6%, 1.7 ±â€¯3.5%, -2.5 ±â€¯6.7%, -0.2 ±â€¯4.9%, -1.2 ±â€¯3.6%, -2.8 ±â€¯7.3%, and -2.1 ±â€¯5.7%, respectively. CONCLUSIONS: Our optimization workflow obtained using DoE for various field configurations provided the same or slightly superior plan quality compared with that created by experts. This process is feasible for clinical practice and will efficiently improve treatment quality while removing the influence of the planner's experience.