Quality Assurance Plan Research Articles

ACPSEM position paper: pre-treatment patient specific plan checks and quality assurance in radiation oncology.

The Australasian College of Physical Scientists and Engineers in Medicine (ACPSEM) has not previously made recommendations outlining the requirements for physics plan checks in Australia and New Zealand. A recent workforce modelling exercise, undertaken by the ACPSEM, revealed that the workload of a clinical radiation oncology medical physicist can comprise of up to 50% patient specific quality assurance activities. Therefore, in 2022 the ACPSEM Radiation Oncology Specialty Group (ROSG) set up a working group to address this issue. This position paper authored by ROSG endorses the recommendations of the American Association of Physicists in Medicine (AAPM) Task Group 218, 219 and 275 reports with some contextualisation for the Australia and New Zealand settings. A few recommendations from other sources are also endorsed to complete the position.

Perencanaan Jaminan Mutu Sekolah Dasar Islam Terpadu

The achievement of educational goals is marked by the quality of education. The quality of education is considered good when the national education goals and the goals of Islamic education are achieved. The planning stage is crucial, especially in relation to ensuring the quality of education. This study analyzes the quality assurance planning of the Terpadu Islamic elementary school, especially those within the Integrated Islamic School Network (JSIT) in Indonesia. This research uses a qualitative approach in the form of literature study and observation. The results obtained show that the Ummul Quro Bogor Integrated Islamic Elementary School has carried out quality assurance planning in accordance with national education standards and the standards of the Integrated Islamic School Network (JSIT).

Single isocenter versus dual isocenter treatment using flattening filter-free and jaw-tracking volumetrically modulated arc therapy for boot-shaped lung cancer: Evaluation of dosimetric and feasibility.

To determine whether a dual-isocenter volumetrically modulated arc therapy (VMAT) technique results in lower normal pulmonary dosage compared to a traditional single isocenter technique for boot-shaped lung cancer. A cohort of 15 patients with advanced peripheral or central lung cancer who had metastases in the mediastinum and supraclavicular lymph nodes was randomly selected for this retrospective study. VMAT plans were generated for each patient using two different beam alignment techniques with the 6-MV flattening filter-free (FFF) photon beam: single-isocenter jaw-tracking VMAT based on the Varian TrueBeam linear accelerator (S-TV), and dual-isocenter VMAT based on both TrueBeam (D-TV) and Halcyon linear accelerator (D-HV). For all 45 treatment plans, planning target volume (PTV) dose coverage, conformity/homogeneity index (CI/HI), mean heart dose (MHD), mean lung dose (MLD) and the total lung tissue receiving 5, 20, 30Gy (V5, V20, V30) were evaluated. The monitor units (MUs), delivery time, and plan quality assurance (QA) results were recorded. The quality of the objectives of the three plans was comparable to each other. In comparison with S-TV, D-TV and D-HV improved the CI and HI of the PTV (p<0.05). The MLD was 13.84±1.44Gy (mean±SD) for D-TV, 14.22±1.30Gy and 14.16±1.42Gy for S-TV and D-HV, respectively. Lungs-V5Gy was 50.78±6.24%, 52.00±7.32% and 53.36±8.48%, Lungs-V20Gy was 23.72±2.27%, 26.18±2.86% and 24.96±3.09%, Lungs-V30Gy was 15.69±1.76%, 17.20±1.72% and 16.52±2.07%. Compared to S-TV, D-TV provided statistically significant better protection for the total lung, with the exception of the lungs-V5. All plans passed QA according the gamma criteria of 3%/3mm. Taking into account the dosimetric results and published clinical data on radiation-induced pulmonary injury, dual-isocenter jaw-tracking VMAT may be the optimal choice for treating boot-shaped lung cancer.

Dose Calculation Accuracy of Beam Models in RadCalc for a 1.5 T MR-Linac

The purpose of this study is to evaluate RadCalc, an independent dose verification software, for patient-specific quality assurance (PSQA) in online adaptive planning with a magnetic resonance linear accelerator (MR-linac) of a 1.5 T. Version 7.1.4 of RadCalc to introduce the capability to establish a beam model that incorporates MR field characteristics. A total of six models were established, with one using manufacturer-provided data and the others differing in percentage depth dose (PDD) data sources. Overall, two models utilized PDD data from the treatment planning system (TPS), and three used commissioned PDD data from gantry angles of 0° and 270°. Simple tests on a virtual water phantom assessed dose-calculation accuracy, revealing percentage differences ranging from −0.5% to −20.6%. Excluding models with significant differences, clinical tests on 575 adaptive plans (prostate, liver, and breast) showed percentage differences of −0.51%, 1.12%, and 4.10%, respectively. The doses calculated using RadCalc demonstrated similar trends to those of the PSQA-based measurements. The newly released version of RadCalc enables beam modeling that considers the characteristics of the 1.5 T magnetic field. The accuracy of the software in calculating doses at 1.5 T magnetic fields has been verified, thereby making it a reliable and effective tool for PSQA in adaptive plans.

Evaluation of 2D ion chamber arrays for patient specific quality assurance using a static phantom at a 0.35 T MR-Linac

IntroductionPatient specific quality assurance (QA) in MR-Linacs can be performed with MR-compatible ion chamber arrays. However, the presence of a static magnetic field can alter the angular response of such arrays substantially. This works investigates the suitability of two ion chamber arrays, an air-filled and a liquid-filled array, for patient specific QA at a 0.35 T MR-Linac using a static phantom. MethodsIn order to study the angular response, the two arrays were placed in a static, solid phantom and irradiated with 9.96 × 9.96 cm2 fields every 10° beam angle at a 0.35 T MR-Linac. Measurements were compared to the TPS calculated dose in terms of gamma passing rate and relative dose to the central chamber. 20 patient specific quality assurance plans were measured using the liquid-filled array. ResultsThe air-filled array showed asymmetric angular response changes of central chamber dose of up to 18% and down to local 3 mm / 3% gamma rates of 20%, while only minor differences within 3% (excluding parallel irradiation and beams through the couch edges) were found for the liquid-filled ion chamber array without rotating the phantom. Patient plan QA using the liquid-filled array yielded a median local 3 mm / 3% 3D gamma passing rate of 99.8% (range 96.9%–100%) ConclusionA liquid-filled ionization chamber array in combination with a static phantom can be used for efficient patient specific plan QA in a single measurement set-up in a 0.35 T MR-Linac, while the air-filled ion chamber array phantom shows large angular response changes and has its limitations regarding patient specific QA measurements

Challenges of Quality Assurance in Early Planning and Ramp Up of Production Facilities - Potentials of Planning Automation via Virtual Engineering

The use of Virtual Engineering is getting more and more common in production facility planning in the automotive sector. However, there are still discrepancies between the reality and the software generated models. That is why the process of quality assurance is crucial. This article aims at determining the current challenges in the planning process of production facilities, which are to be processed by means of Virtual Engineering, and limitations that occur in practice during the execution of Virtual Engineering in digital production facility engineering. We strive to maximize the utilization of underlying potentials by seeking out solutions. Our approach involves conducting a systematic literature review as well as seven expert interviews to investigate the current practical implementation of Virtual Engineering at Original Equipment Manufacturers (OEMs). Additionally, we aim to validate the influencing variables obtained from the literature in relation to the challenges, limitations, and potentials of Virtual Engineering. We show that there are current challenges in the planning process of production facilities, where the areas of functionality, equipment selection, safety, accessibility, and key performance indicators should already be secured in a virtual environment by means of Virtual Engineering. Our findings reveal significant limiting factors in the application of Virtual Engineering as quality assurance. This limitations arises as a result of the reliance on manual activities, which ultimately hinders progress in terms of time efficiency, quality assurance, and cost-effectiveness. In order to optimize this, we show that the potential of standardization of the work steps is required to define an automation that fully exploits the potentials of Virtual Engineering.

Delta4-based Dosimetric Error Detection in Volumetric-modulated Arc Therapy: Clinical Significance and Implications.

Volumetric-modulated arc therapy (VMAT) is an efficient method of administering intensity-modulated radiotherapy beams. The Delta4 device was employed to examine patient data. The utility of the Delta4 device in identifying errors for patient-specific quality assurance of VMAT plans was studied in this research. Intentional errors were purposely created in the collimator rotation, gantry rotation, multileaf collimator (MLC) position displacement, and increase in the number of monitor units (MU). The results show that when the characteristics of the treatment plans were changed, the gamma passing rate (GPR) decreased. The largest percentage of erroneous detection was seen in the increasing number of MU, with a GPR ranging from 41 to 92. Gamma analysis was used to compare the dose distributions of the original and intentional error designs using the 2%/2 mm criteria. The percentage of dose errors (DEs) in the dose-volume histogram (DVH) was also analyzed, and the statistical association was assessed using logistic regression. A modest association (Pearson's R-values: 0.12-0.67) was seen between the DE and GPR in all intentional plans. The findings indicated a moderate association between DVH and GPR. The data reveal that Delta4 is effective in detecting mistakes in treatment regimens for head-and-neck cancer as well as lung cancer. The study results also imply that Delta4 can detect errors in VMAT plans, depending on the details of the defects and the treatment plans employed.

When field seam test results sail, yet laboratory test results pass: A practical approach to understanding why

A standard quality control (QC) and quality assurance (QA) plan during installation of an unreinforced polyethylene geomembrane often requires destructive testing of the production welded seams at regular intervals. This testing is different from “trial weld testing.” Prior to welding of the seams, within the project footprint, a “trial weld” is fabricated on site as a proficiency test. The procedure utilizes project specific geomembrane material, environmental conditions, welding technician and equipment intended to perform the seaming of geomembrane installed as part of the containment system. There are cases where on-site trial welds become difficult to pass, even when experienced crews, quality equipment and materials are present. The difficulty can lead to a scenario where trial welds fail and destructive seam tests pass when sent to a 3rd party lab for testing, (this is a nightmare for geomembrane installers). Variables are numerous and often difficult to identify between these two different testing scenarios. However, this paper will show results and insightful data that will give rationale for the discrepancy.

The Impact of Dose Calibration Point Differences Between the Treatment Plan and Patient-specific Quality Assurance in Passive Scattering Proton Beam Therapy.

Passive scattering proton beam therapy (PSPT) is performed by taking actual measurements of all pre-designated fields in a treatment plan followed by appropriate adjustments to the prescribed dose. For this reason, it is necessary to ensure precision management of the measurements (patient-specific calibration) in the administration of a planned dose. Therefore, this study investigated the impact on dose distribution in treatment planning when the patient calibration point differs from the normalized point in a treatment plan. A total of 16 cases were selected, where the patient calibration point and normalized point did not match, and the normalized point used in the treatment plan was changed to the patient calibration point using a treatment planning system (VQA ver. 2.01, HITACHI). At this point, the displacement of the relative dose at the isocenter was estimated as an error owing to the difference compared to the patient calibration point. Overall, the error was within the range of ±1.5%, with the exception of orbit cases. Calibrated points also tended to be lower than the normalized points in the treatment plan. In terms of treatment sites, a greater deviation was observed for head cases. Cases with a large deviation in sites other than the head were attributed to poor flatness within the radiation field owing to a narrower opening of the patient collimator. Dose measurement errors in PSPT due to differing calibration points were generally within ±1.5%, with higher deviations observed in head treatments because of complex structures and narrow collimator openings. A γ analysis for significant deviations showed a 98.7% passing rate, suggesting limited overall impact. It is important to select stable calibration points in dosimetry to ensure high precision in dose administration, particularly in complex treatment areas.

Comparison of Rapid Arc and Intensity Modulated Radiotherapy in a True Beam Linear Accelerator for 6 MV: Application of AAPM TG‐119 tests in treatment planning and quality assurance

AbstractObjectiveWe developed rapid arc (RA) and intensity‐modulated radiotherapy (IMRT) plans based on the American Association of Physicists in Medicine Task Group‐119 (AAPM TG‐119) proposals and compared the planning and quality assurance results.Materials and MethodsTwo treatment plans were used for each study patient: one using 7–9 IMRT fields and the other using the two full‐arc RA methods. Plan optimization and dose calculations were performed using 6 MV photons and EclipseTM with the anisotropic analytical algorithm (AAA). Task group (TG)‐119 described the planning objectives used to evaluate the treatment plans generated for this study. Point dose, planar fluence measurement, and trajectory log file analysis are used for quality assurance.ResultsThe conformity index (CI) for treatment plans varied between 0.76–0.91 when using IMRT and 0.75–0.93 when using RA. The homogeneity index (HI) was approximately 0.10–0.24 (IMRT) and 0.07–0.23 (RA). The ratio of the total number of monitor units required for IMRT to that required for RA was between 0.87 and 2.14. The treatment log files exhibited higher gamma passing with RA than with IMRT.ConclusionRA plans were more effective than IMRT plans in achieving the test goals. Point dose measurements and electronic portal imaging device (EPID)‐based planar fluence measurements showed statistically insignificant differences between the IMRT and RA plan quality assurance (QA) results. However, the trajectory log file analysis exhibited better gamma‐passing results for RA than for IMRT.

The Urgency Of Quality Assurance For Educational Customers

The aim of this research is to find out how urgent quality assurance is for education customers. This research method uses literature study by collecting from various reading sources such as e-books, journals or examining online literature. The research results found that Quality Assurance is a quality guarantee offered by an institution. In line with the times, quality assurance has also been developed by educational institutions. Quality assurance carried out by educational institutions is more oriented towards guaranteeing the quality of education offered to educational customers. There are several efforts to improve the quality of education through quality assurance, including good quality assurance planning, implementation of quality assurance planning and evaluation of quality assurance carried out by educational institutions

The performance of a new type accelerator uRT-linac 506c evaluated by a quality assurance automation system.

The purpose of this study was to evaluate the performance of our quality assurance (QA) automation system and to evaluate the machine performance of a new type linear accelerator uRT-linac 506c within 6months using this system. This QA automation system consists of a hollow cylindrical phantom with 18 steel balls in the phantom surface and an analysis software to process electronic portal imaging device (EPID) measurement image data and report the results. The performance of the QA automation system was evaluated by the tests of repeatability, archivable precision, detectability of introduced errors, and the impact of set-up errors on QA results. The performance of this linac was evaluated by 31 items using this QA system over 6months. This QA system was able to automatically deliver QA plan, EPID image acquisition, and automatic analysis. All images acquiring and analysis took approximately 4.6min per energy. The preset error of 0.1mm in multi-leaf collimator (MLC) leaf were detected as 0.12±0.01mm for Bank A and 0.10±0.01mm in Bank B. The 2mm setup error was detected as -1.95±0.01mm, -2.02±0.01mm, 2.01±0.01mm for X, Y, Z directions, respectively. And data from the tests of repeatability and detectability of introduced errors showed the standard deviation were all within 0.1mm and 0.1°. and data of the machine performance were all within the tolerance specified by AAPM TG-142. The QA automation system has high precision and good performance, and it can improve the QA efficiency. The performance of the new accelerator has also performed very well during the testing period.

Feasibility study of adaptive radiotherapy with Ethos for breast cancer.

The aim of this study was to assess the feasibility of online adaptive radiotherapy with Ethos for breast cancer. This retrospective study included 20 breast cancer patients previously treated with TrueBeam. All had undergone breast surgery for different indications (right/left, lumpectomy/mastectomy) and were evenly divided between these four cases, with five extended cone beam computed tomography (CBCT) scans per patient. The dataset was used in an Ethos emulator to test the full adaptive workflow. The contours generated by artificial intelligence (AI) for the influencers (left and right breasts and lungs, heart) and elastic or rigid propagation for the target volumes (internal mammary chain (IMC) and clavicular lymph nodes (CLNs)) were compared to the initial contours delineated by the physician using two metrics: Dice similarity coefficient (DICE) and Hausdorff 95% distance (HD95). The repeatability of influencer generation was investigated. The times taken by the emulator to generate contours, optimize plans, and calculate doses were recorded. The quality of the scheduled and adapted plans generated by Ethos was assessed using planning target volume (PTV) coverage, homogeneity indices (HIs), and doses to organs at risk (OARs) via dose-volume histogram (DVH) metrics. Quality assurance (QA) of the treatment plans was performed using an independent portal dosimetry tool (EpiQA) and gamma index. On average, the DICE for the influencers was greater than 0.9. Contours resulting from rigid propagation had a higher DICE and a lower HD95 than those resulting from elastic deformation but remained below the values obtained for the influencers: DICE values were 0.79 ± 0.11 and 0.46 ± 0.17 for the CLN and IMC, respectively. Regarding the repeatability of the influencer segmentation, the DICE was close to 1, and the mean HD95 was strictly less than 0.15mm. The mean time was 73 ± 4 s for contour generation per AI and 80 ± 9 s for propagations. The average time was 53 ± 3 s for dose calculation and 125 ± 9 s for plan optimization. A dosimetric comparison of scheduled and adapted plans showed a significant difference in PTV coverage: dose received by 95% of the volume (D95%) values were higher and closer to the prescribed doses for adapted plans. Doses to organs at risk were similar. The average gamma index for quality assurance of adapted plans was 99.93 ± 0.38 for a 3%/3mm criterion. This study comprehensively evaluated the Ethos® adaptive workflow for breast cancer and its potential technical limitations. Although the results demonstrated the high accuracy of AI segmentation and the superiority of adapted plans in terms of target volume coverage, a medical assessment is still required.

Knowledge-based model building for treatment planning for prostate cancer using commercial treatment planning quality assurance software tools.

This study devised a method to efficiently launch the RapidPlan model for volumetric-modulated arc therapy for prostate cancer in small- and medium-sized facilities using high-quality treatment plans with the PlanIQ software as a reference. Treatment plans were generated for 30 patients with prostate cancer to construct the RapidPlan model using PlanIQ as a reference. In the context of PlanIQ-referenced treatment planning, treatment plans were developed, such that the feasibility dose-volume histogram of each organ-at-risk fell within F ≤ 0.1. For validation of the RapidPlan model, treatment plans were formulated for 20 patients using both RapidPlan and PlanIQ, and the differences were evaluated. The results of RapidPlan model validity assessment revealed that the RapidPlan-produced treatment plans exhibited higher quality in 11 of 20 patients. No significant differences were found between the treatment plans. In conclusion, high-quality treatment plans formulated using PlanIQ as reference facilitated efficient implementation of RapidPlan modeling.

A robust evaluation of 49 high-dose-rate prostate brachytherapy treatment plans including all major uncertainties.

Uncertainties in radiotherapy cause deviation from the planned dose distribution and may result in delivering a treatment that fails to meet clinical objectives. The impact of uncertainties is unique to the patient anatomy and the needle locations in HDR prostate brachytherapy. Evaluating this impact during treatment planning is not common practice, relying on margins around the target or organs-at-risk to account for uncertainties. A robust evaluation framework for HDR prostate brachytherapy treatment plans was evaluated on 49 patient plans, measuring the range of possible dosimetric outcomes to the patient due to 14 major uncertainties. Patient plans were evaluated for their robustness to uncertainties by simulating probable uncertainty scenarios. Five-thousand probabilistic and 1943 worst-case scenarios per patient were simulated by changing the position and size of structures and length of dwell times from their nominal values. For each uncertainty scenario, the prostate D90 and maximum doses to the urethra, D0.01cc , and rectum, D0.1cc , were calculated. The D90 was an average 1.16±0.51% (mean±SD) below nominal values for the probabilistic scenarios; the D0.01cc metric was 2.24±0.90% higher; and D0.1cc was greater by 0.48±0.30%. The D0.01cc and D90 metrics were more sensitive to uncertainties than D0.1cc , with a median of 79.0% and 84.9% of probabilistic scenarios passing the constraints, compared to 96.5%. The median pass-rate for scenarios that passed all three metrics simultaneously was 63.4%. Assessing treatment plan robustness improves plan quality assurance, is achievable in less than 1-min, and identifies treatment plans with poor robustness, allowing re-optimization before delivery.

A dialectical system framework for enhancing the modular construction supply chain

Modular construction (MC) is an innovative approach to enhancing productivity and sustainability, but its complex and dynamic supply chain is challenging. This paper develops and validates a dialectical system framework for understanding and enhancing the MC supply chain, through the combination of a literature review, interviews with 59 professionals and a case study of modular integrated construction development in Hong Kong, China. The developed framework encompasses seven key components of the MC supply chain technical system – namely, pre-approval, building plan pre-submission, detailed design, module production, logistics, module installation, and quality assurance and quality control – considering various stakeholders embedded in broader contexts. The case study validates the framework as a theoretical lens to comprehend the complex MC supply chain and as a novel analytical model to enhance the MC supply chain from different stakeholder and contextual perspectives. Future research directions are suggested regarding strategy establishment, performance measurement and policy road map development to achieve a mature and sustainable supply chain.

Differences between professionals in treatment planning for patients with stage III lung cancer using treatment-planning QA software.

The quality of treatment planning for stage III non-small cell lung cancer varies within and between facilities due to the different professions involved in planning. Dose estimation parameters were calculated using a feasibility dose-volume histogram (FDVH) implemented in the treatment planning quality assurance software PlanIQ. This study aimed to evaluate differences in treatment planning between occupations using manual FDVH-referenced treatment planning to identify their characteristics. The study included ten patients with stage III non-small cell lung cancer, and volumetric-modulated arc therapy was used as the treatment planning technique. Fifteen planners, comprising five radiation oncologists, five medical physicists, and five radiological technologists, developed treatment strategies after referring to the FDVH. Medical physicists had a higher mean dose at D98% of the planning target volume (PTV) and a lower mean dose at D2% of the PTV than those in other occupations. Medical physicists had the lowest irradiation lung volumes (V5 Gy and V13 Gy) compared to other professions, and radiation oncologists had the lowest V20 Gy and mean lung dose. Radiological technologists had the highest irradiation volumes for dose constraints at all indexes on the normal lung volume. The quality of the treatment plans developed in this study differed between occupations due to their background expertise, even when an FDVH was used as a reference. Therefore, discussing and sharing knowledge and treatment planning techniques among professionals is essential to determine the optimal treatment plan for each facility and patient.

Implementation of a Patient Specific QA Protocol for a Novel Dedicated Stereotactic Radiosurgery Linear Accelerator

To describe the patient specific QA (Quality Assurance) protocol implemented for plan integrity verification, in stereotactic radiosurgery (SRS) and radiotherapy (SRT) treatments performed on a novel dedicated frameless image guided radiosurgery system using conical collimator, and a combination of yoked gimbals to cover near2pi solid angle. Startup and commissioning results obtained for planned vs measured dose distributions with several detectors are presented. The patient specific QA protocol includes: A) Review of the approved plan based on AAPM TG 275 report recommendations B) 1D verification using small volume ionization chambers inside an anthropomorphic head phantom C) 2D dose verification of a coronal and sagittal dose plan performed with high resolution 2D Array and respective phantom apparatus D) In-house independent MU (Monitor Unit) calculation using the same formalism of the vendor TPS (Treatment Planning Systems) RayTracing dose-calculation algorithm E) Monte Carlo based secondary dose check & plan QA. A retrospective analysis of the results of the first 15 patients treated is presented, focusing on plan complexity vs QA results. The 1D results obtained for the 15 SRS plans were within ±5% for all reported cases, with a mean percent difference of -1,25%, confirming an overall good agreement and, as expected, a partial volume effect in plans with smaller collimators. For the 2D dose verifications, with a 10% dose threshold, gamma passing rates of 97,5% (coronal) and 96,78 (sagittal) with 3% 1mm criteria, 95,1% (coronal) and 94,3% (sagittal) for 2% 1 mm criteria and 90,7% (coronal) and 90,0% (sagittal) for 1% 1mm criteria were obtained. Moreover, the results showed a correlation between lesion volume or number of collimators used with gamma passing rates. All MU verification results were within ±0,3% and provides an efficient risk mitigation approach for this new delivery technique. The verification results of the first 15 treated SRS plans confirmed point dose and planar measurements in agreement with TPS calculations, with superior results for planes with smaller lesions and fewer collimators. This also represents an integral validation of the image-based alignment system and fine treatment couch movements, as treatments are intrinsically multi-isocentric.

A Feasibility Study of Dose Band Prediction in Radiotherapy: Predicting a Dose Spectrum

Current deep learning-based dose prediction methods can only predict a specific dose distribution. If the predicted dose is inaccurate, no more options can be selected. We proposed a novel dose prediction method named dose band prediction, which outcomes a spectrum of predicted dose distribution for planning and quality assurance (QA). Upper-Band and Lower-Band losses were involved in 3D convolution neural networks to establish the Upper-Band Network (UBN) and Lower-Band Network (LBN). Each voxel's ideal dose spectrum (dose band) was defined by the maximum/minimum rational dose predicted by UBN/LBN. 130 NPC cases with Tomotherapy (dataset 1), 49 cervix cases with IMRT (dataset 2) and 43 cervix cases with VMAT (dataset 3) were enrolled to establish and evaluate our dose band prediction method. The dose band prediction method can successfully predict a spectrum of doses. Upper-Band/Lower-Band presents maximum/minimum rational dose; Middle-Line presents the average of Upper-Band and Lower-Band. The clinical implement dose was used as the reference dose. We evaluated the maximum interval between the reference and Upper-Band/Middle-Line/Lower-Band doses, and the percentage dose difference was used as the evaluation method. The differences in PTV for Upper-Band, Middle-Line and Lower-Band in dataset 1 were within 2.47%, 0.54%, and 2.8%; in dataset 2, they were within 0.37%, 1.15%, and 2.69%; in dataset 3, they were within 0.96%, 0.35%, and 1.66%. The mean difference of OARs for the Upper-Band, Middle-Line and Lower-Band in dataset 1 were within 8.13%, 4.97%, and 8.19%; in dataset 2, they were within 8.8%, 4.48%, and 5.52%; in dataset 3, they were within 4.01%, 3.13%, and 5.79% (shown in Table 1). Dose Band prediction achieved high-accuracy dose prediction by the Middle-Line. More importantly, the Upper-Band/Lower-Band provided a spectrum of possible rational doses. Our Dose Band prediction method is based on a specific loss function, so it can easily be applied in various network and patient cases. Dose Band prediction towards a more robust plan QA and planning assistance. Table 1. The maximum interval of doses (percentage dose difference, %).

A 3D Deep Convolutional Neural Network for Dose Distribution Prediction of Intracavitary HDR Brachytherapy of Cervical Cancer

The purpose of this work is to develop a voxel-wise dose prediction model using convolutional neural network (CNN) for cervical cancer high-dose-rate (HDR) intracavitary brachytherapy treatment planning with tandem-and-ovoid (T&O) or tandem-and-ring (T&R) applicators. A 3D U-NET CNN was implemented to generate voxel-wise dose predictions based on high-risk clinical target volume (HRCTV) and organs at risk (OAR) contour information. A multi-institutional cohort of 77 retrospective clinical HDR brachytherapy plans treated to a prescription dose in the range of 4.8-7.0 Gy/fx was used in this study. Those plans were randomly divided into 60%/20%/20% as training, validating, and testing cohorts. Data augmentation techniques were implemented in the training and validation cohort data to increase the number of plans to 252. The model was trained using the mean-squared loss function, Adam optimization algorithm, a learning rate of 0.001, 250 epochs, and a batch size of 8. The model performance was evaluated on the testing dataset by analyzing the outcomes in terms of mean dose values and derived dose-volume-histogram (DVH) indices from 3D dose distributions and comparing the generated dose distributions against the ground-truth dose distributions using dose statistics and clinically meaningful dosimetric indices. The proposed 3D U-Net model showed competitive accuracy in predicting 3D dose distributions that closely resemble the ground truth dose distributions. The average value of mean absolute error was 0.108±3.617 Gy for HRCTV, 0.074±1.315 Gy for bladder, 0.093±0.981 Gy for rectum, and 0.035±2.789 Gy for sigmoid. The median absolute error was 1.264 Gy for HRCTV, 0. 441 Gy for the bladder, 0.018 Gy for rectum, and 0.019 Gy for sigmoid. Our results showed that the predicted mean D2cc OAR doses in the bladder, rectum, and sigmoid were 4.02±0.93, 3.77±0.57 and 3.36±0.82 Gy in comparison to 4.35±0.98, 3.99±0.53, 3.75±0.83 Gy in clinical plans, respectively, and the mean individual differences (ΔD2cc) were 0.33±0.41, 0.22±0.16 and 0.39±0.45 Gy, respectively. The predicted D90 of the HRCTV was 6.72±0.99 Gy in comparison with 6.83±1.72 Gy in clinical plans. Results indicate a good potential to predict useful dosimetric indices and thus would facilitate an improvement in brachytherapy treatment planning workflow. The proposed model needs less than 5 seconds to predict a full 3D dose distribution of 64 × 64 × 64 voxels for any new patient plan, thus making it sufficient for near real-time applications and aid in decision-making in clinic. The proposed model can be used to predict 3D dose distributions for near real-time decision-making, before planning, for planning quality assurance, and for guiding future automated planning for improved plan consistency, quality, and planning efficiency.