Nutrition Education Models for Patients With Chronic Kidney Disease.

Nutrition is an integral component in the management of chronic kidney disease (CKD), and kidney health professionals play a crucial role in educating patients on dietary interventions for CKD. Several dietary modifications are indicated for CKD that require frequent adaptations with CKD progression and with underlying metabolic disturbances. However, poor adherence to dietary interventions is not uncommon among patients with CKD. An effective education program on nutrition intervention consists of providing knowledge and developing skills that are necessary to support behavioral change. The application of theoretical models of behavioral change such as social cognitive theory and the transtheoretical model in nutrition intervention has been reported to be effective in promoting changes in dietary habits. This review summarizes the evidence supporting the application of theoretical models as strategies to enhance nutrition education for patients with CKD. In addition, digital technologies are gaining interest in empowering patients and facilitating nutrition management in patients with CKD. This review also examines the applications of the latest digital technologies guided by behavioral theory in facilitating patients' changes in dietary intake patterns and lifestyle habits.

Predictive quality assurance for linear accelerator target failure using statistical process control.

The performance of a linear accelerator (Linac) depends on the integrity of its x-ray target. The sudden failure of its target not only breaks down the Linac but also could contribute significant disruptions to patient care. This work is to develop a predicative quality assurance (QA) method using Statistical Process Control (SPC) and AutoRegressive Integrated Moving Average (ARIMA) modeling to identify the risk of target failure before it occurs. In the past years, we observed two incidents of target failure among our Linacs. Retrospectively, we collected past daily QA data (from both open fields and enhanced dynamic wedge (EDW) measurements) and analyzed its historical trend using methods of SPC and ARIMA. SPC is a technique that monitors process performance based on statistical analysis. ARIMA is a time-series forecasting algorithm that can be used to estimate future values based on its past pattern. Both have been evaluated for predictive QA in radiotherapy. Application of SPC on open beam QA data would not yield an early warning signal to the pending target failures. However, when the same SPC methodology applies to EDW measurements, the control limits were breached a couple of weeks before the target failed. EDW mechanism introduces nonuniform magnification factors over its wedge-directed beam profiles and is responsible for the sensitivity of its profile to changing beam properties induced by a degrading target. Further extension of the warning period may be possible by using ARIMA modeling. Predicative QA for EDW daily data using SPC and ARIMA methods may provide an early QA warning to incoming Linac target failure.

Technical note: Energy dependence of the Gafchromic EBT4 film: Dose-response curves for 70 kV, 6 MV, 6 MV FFF, 10 MV FFF, and 15 MV x-ray beams.

BACKGROUND: EBT4 was newly released for radiotherapy quality assurance to improve the signal-to-noise ratio in radiochromic film dosimetry. It is important to know its dose-response characteristics before its use in the clinic. PURPOSE: This study aims to investigate and compare the dose-response curves of the Gafchromic EBT4 film for megavoltage and kilovoltage x-ray beams with different dose levels, scanning spatial resolutions, and sizes of region of interest (ROI). METHODS: EBT4 film (Lot#07052201) calibration strips (3.5 × 20 cm2 ) were exposed to a 10×10 cm2 open field at doses of 0, 63, 125, 500, 750, 1000 cGy using 6 MV photon beam. EBT4 film strips from the same lot were then exposed to each x-ray beam (6 MV, 6 MV FFF, 10 MV FFF, 15 MV, and 70 kV) at six dose values (50, 100, 300, 600, 800, 1000 cGy). A full sheet (25 × 20 cm2 ) of EBT4 film was irradiated at each energy with 300 cGy for profile comparison with the treatment planning calculation. At two different spatial resolutions of 72 and 300 dpi, each film piece was scanned three consecutive times in the center of an Epson 10000XL flatbed scanner in 48-bit color. The scanned images were analyzed using FilmQA Pro. For each scanned image, an ROI of 2 × 2 cm2 at the field center was selected to obtain the average pixel value with its standard deviation in the ROI. An additional ROI of 1 cm diameter circle was also used to evaluate the impact of ROI shape and size, especially for FFF beams. The dose value, average dose-response value, and associated uncertainty were determined for each energy and relative responses were analyzed. The Student's t-test was performed to evaluate the statistical significance of the dose-response values with different color channels, ROI shapes, and spatial resolutions. RESULTS: The dose-response curves for the five x-ray energies were compared in three color channels. Weak energy dependence was found among the megavoltage beams. No significant differences (average ∼1.1%) were observed for all doses in this study among 6 MV, 6 MV FFF, 10 MV FFF, and 15 MV beams, regardless of spatial resolution and color channel. However, a statistically significant difference in dose-response was observed up to 12% between 70 kV and 6 MV beams. CONCLUSIONS: The dose-response curves for Gafchromic EBT4 films were nearly independent of the energy of the photon beams among 6 MV, 6 MV FFF, 10 MV FFF, and 15 MV. For very low-energy photons (e.g., 70 kV), a separate calibration from the same low-energy x-ray is necessary.

Comparative study of SRS end-to-end QA processes of a diode array device and an anthropomorphic phantom loaded with GafChromic XD film.

PURPOSE: End-to-end testing (E2E) is a necessary process for assessing the readiness of the stereotactic radiosurgery (SRS) program and annual QA of an SRS system according to the AAPM MPPG 9a. This study investigates the differences between using a new SRS MapCHECK (SRSMC) system and an anthropomorphic phantom film-based system in a large network with different SRS delivery techniques. METHODS AND MATERIALS: Three SRS capable Linacs (Varian Medical Systems, Palo Alto, CA) at three different regional sites were chosen to represent a hospital network, a Trilogy with an M120 multi-leaf collimator (MLC), a TrueBeam with an M120 MLC, and a TrueBeam Stx with an HD120 MLC. An anthropomorphic STEEV phantom (CIRS, Norfolk, VA) and a phantom/diode array: StereoPHAN/SRSMC (Sun Nuclear, Melbourne, FL) were CT scanned at each site. The new STV-PHANTOM EBT-XD films (Ashland, Bridgewater, NJ) were used. Six plans with various complexities were measured with both films and SRSMC in the StereoPHAN to establish their dosimetric correlations. Three SRS cranial plans with a total of sixteen fields using dynamic conformal arc and volumetric-modulated arc therapy, with 1-4 targets, were planned with Eclipse v15.5 treatment planning system (TPS) using a custom SRS beam model for each machine. The dosimetric and localization accuracy were compared. The time of analysis for the two systems by three teams of physicists was also compared to assess the throughput efficiency. RESULTS: The correlations between films and SRSMC were found to be 0.84 (p = 0.03) and 0.16 (p = 0.76) for γ (3%, 1 mm) and γ (3%, 2 mm), respectively. With film, the local dose differences (ΔD) relative to the average dose within the 50% isodose line from the three sites were found to be -3.2%-3.7%. The maximum localization errors (Elocal ) were found to be within 0.5 ± 0.2 mm. With SRSMC, the ΔD was found to be within 5% of the TPS calculation. Elocal were found to be within 0.7 to 1.1 ± 0.4 mm for TrueBeam and Trilogy, respectively. Comparing with film, an additional uncertainty of 0.7 mm was found with SRSMC. The delivery and analysis times were found to be 6 and 2 h for film and SRSMC, respectively. CONCLUSIONS: The SRS MapCHECK agrees dosimetrically with the films within measurement uncertainties. However, film dosimetry shows superior sub-millimeter localization resolving power for the MPPG 9a implementation.

Impact of a Centralized Database System on Radiation Therapy Quality Assurance Management at a Large Health Care Network: 5 Years' Experience.

PURPOSE: This study reports the impact of using a centralized database system for major equipment quality assurance (QA) at a large institution. METHODS AND MATERIALS: A centralized database system has been implemented for radiation therapy machine QA in our institution at 6 campuses with 11 computed tomographies and 22 linear accelerators (LINACs). The database system was customized to manage monthly and annual computed tomography and LINAC QA. This includes providing the same set of QA procedures across the enterprise, digitally storing all measurement records, and generating trend analyses. Compared with conventional methods (ie, paper forms), the effectiveness of the database system was quantified by changes in the compliance of QA tests and perceptions of staff to the efficiency of data retrieval and analyses. An anonymized questionnaire was provided to physicists enterprise-wide to assess workflow changes. RESULTS: With the implementation of the database system, the compliance of QA test completion improved from 80% to >99% for the entire institution. This resonates with the 56% of physicists who found the database system helpful in guiding them through QA, and 25% of physicists found the contrary, and 19% reported no difference (n = 16). Meanwhile, 40% of physicists reported longer times needed to record data using the database system compared with conventional methods, and another 40% suggested otherwise. In addition, 87% and 80% found the database more efficient to analyze and retrieve previous data, respectively. This was also reflected by the shorter time taken to generate year-end QA statistics using the software (5 vs 30 min per LINAC). Overall, 94% of physicists preferred the centralized database system over conventional methods and endorsed continued use of the system. CONCLUSIONS: A centralized database system is useful and can improve the effectiveness and efficiency of QA management in a large institution. With consistent data collection and proper data storage using a database, high-quality data can be obtained for failure modes and effects analyses as per TG 100.

An evaluation of the use of EBT-XD film for SRS/SBRT commissioning of a 1.5 Tesla MR-Linac system.

PURPOSE: The goal of this study was to evaluate the use of EBT-XD film for SRS/SBRT commissioning in a 1.5T hybrid MR-Linac (MRL). METHOD: The output factors (St), from 1x1, 2x2, 3x3 cm2, were measured with film in solid water. The results were compared with (1) the measurements by a PTW diamond detector (CVD) and an Exradin® A26MR ion chamber in 3D water phantom; (2) Monte Carlo calculation by Monaco TPS (MTPS) in water. The inline (IN) and crossline (CR) profiles, measured by films and the CVD, were also compared. An SRS plan with two targets was created in MTPS and was measured with EBT-XD film in a StereoPHANTM phantom serving as an end-to-end test. The 3x3 cm2 was used for film calibration with doses ranging from 0 to 28 Gy. Water was added to the phantom-film-phantom interface to reduce the electron-return-effect (ERE). Films were calibrated with One-scan-dosimetry protocol. RESULTS: The film St were within 1.2% and 2.2% compared to other detectors and MTPS respectively. At the central B-field induced asymmetric region, films were within 0.6% between the CVD and the MTPS, but 5-8% differences were observed in the 40%-5% gradient region in CR due to ERE. The differences in localization and dose were found to be 0.6 mm and 3.3%. The γ(3%/2mm), γ (5%/2mm), γ (5%/1mm) were 97.7%, 99.3%, 94.6%. CONCLUSIONS: Films can provide accurate dosimetric results under ERE and are valuable for commissioning MRL. Using the One-scan-dosimetry protocol with EBT-XD film for MRL increases accuracy and efficiency in commissioning and QA of SRS/SBRT.

A method for in vivo treatment verification of IMRT and VMAT based on electronic portal imaging device.

BACKGROUND: Intensity-modulated radiation therapy (IMRT) and volume-modulated arc therapy (VMAT) are rather complex treatment techniques and require patient-specific quality assurance procedures. Electronic portal imaging devices (EPID) are increasingly used in the verification of radiation therapy (RT). This work aims to develop a novel model to predict the EPID transmission image (TI) with fluence maps from the RT plan. The predicted TI is compared with the measured TI for in vivo treatment verification. METHODS: The fluence map was extracted from the RT plan and corrections of penumbra, response, global field output, attenuation, and scatter were applied before the TI was calculated. The parameters used in the model were calculated separately for central axis and off-axis points using a series of EPID measurement data. Our model was evaluated using a CIRS thorax phantom and 20 clinical plans (10 IMRT and 10 VMAT) optimized for head and neck, breast, and rectum treatments. RESULTS: Comparisons of the predicted and measured images were carried out using a global gamma analysis of 3%/2 mm (10% threshold) to validate the accuracy of the model. The gamma pass rates for IMRT and VMAT were greater than 97.2% and 94.5% at 3%/2 mm, respectively. CONCLUSION: We have developed an accurate and straightforward EPID-based quality assurance model that can potentially be used for in vivo treatment verification of the IMRT and VMAT delivery.

Risk Management of Clinical Reference Dosimetry of a Large Hospital Network Using Statistical Process Control.

Managing TG-51 reference dosimetry in a large hospital network can be a challenging task. The objectives of this study are to investigate the effectiveness of using Statistical Process Control (SPC) to manage TG-51 workflow in such a network. All the sites in the network performed the annual reference dosimetry in water according to TG-51. These data were used to cross-calibrate the same ion chambers in plastic phantoms for monthly QA output measurements. An energy-specific dimensionless beam quality cross-calibration factor, k q n S W , was derived to monitor the process across multiple sites. The SPC analysis was then performed to obtain the mean, 〈 k q n S W 〉 , standard deviation, σ k , the Upper Control Limit (UCL) and Lower Control Limit (LCL) in each beam. This process was first applied to 15 years of historical data at the main campus to assess the effectiveness of the process. A two-year prospective study including all 30 linear accelerators spread over the main campus and seven satellites in the network followed. The ranges of the control limits (±3σ) were found to be in the range of 1.7% - 2.6% and 3.3% - 4.2% for the main campus and the satellite sites respectively. The wider range in the satellite sites was attributed to variations in the workflow. Standardization of workflow was also found to be effective in narrowing the control limits. The SPC is effective in identifying variations in the workflow and was shown to be an effective tool in managing large network reference dosimetry.

A Review and Analysis of Managing Commonly Seen Implanted Devices for Patients Undergoing Radiation Therapy.

PURPOSE: This review article aims to consolidate information regarding existing and emerging implanted devices used in patients undergoing radiation therapy and to categorize levels of attention needed for each device, including which devices require monitoring throughout treatment. METHODS AND MATERIALS: Based on the collective information from scholar searches, manufacturers' technical reports, and institutional experiences in the past years, commonly present devices in patients with cancer are compiled. This work summarizes cardiac pacemaker, implanted cardiac defibrillator, hepatic pump, intrathecal pain pump, neurostimulator, shunt, loop recorder, and mediport. Three different classifications of implanted devices can be made based on the potential effect of radiation: life-dependent, nonlife-dependent but with adverse effects if overdosed, and devices without electronic circuits. Implanted devices that contain electronic circuits that would be life-dependent or have adverse effects if overdosed, include cardiac pacemakers, implanted cardiac defibrillators, programmable hepatic pumps, pain pumps, neurostimulators, and loop recorders. RESULTS: Dose exposure to these devices need to be calculated or measured in vivo, especially for cardiac implanted devices, and they should be minimized to assure continued healthy functioning. Treatment planning techniques should be chosen to reduce entry, exit and internal scatter dose. Lower energy photon beams should be used to decrease potential neutron contamination. Implanted devices without electronic circuits are less of a concern. If a patient is life-dependent on the implanted device, it is not recommended to treat the patient with proton therapy. CONCLUSIONS: This study reviewed the management of patients with commonly seen implanted devices and summarized a workflow for identifying and planning when a patient has implanted devices. Classifications of implanted devices could help clinicians make proper decisions in regard to patients with specific implanted devices. Lastly, the management of such devices in the era of the pandemic is also discussed in this review article.

Commissioning and clinical implementation of an Autoencoder based Classification-Regression model for VMAT patient-specific QA in a multi-institution scenario.

BACKGROUND AND PURPOSE: To commission and implement an Autoencoder based Classification-Regression (ACLR) model for VMAT patient-specific quality assurance (PSQA) in a multi-institution scenario. MATERIALS AND METHODS: 1835 VMAT plans from seven institutions were collected for the ACLR model commissioning and multi-institutional validation. We established three scenarios to validate the gamma passing rates (GPRs) prediction and classification accuracy with the ACLR model for different delivery equipment, QA devices, and treatment planning systems (TPS). The prediction performance of the ACLR model was evaluated using mean absolute error (MAE) and root mean square error (RMSE). The classification performance was evaluated using sensitivity and specificity. An independent end-to-end test (E2E) and routine QA of the ACLR model were performed to validate the clinical use of the model. RESULTS: For multi-institution validations, the MAEs were 1.30-2.80% and 2.42-4.60% at 3%/3 mm and 3%/2 mm, respectively, and RMSEs were 1.55-2.98% and 2.83-4.95% at 3%/3 mm and 3%/2 mm, respectively, with different delivery equipment, QA devices, and TPS, while the sensitivity was 90% and specificity was 70.1% at 3%/2 mm. For the E2E, the deviations between the predicted and measured results were within 3%, and the model passed the consistency check for clinical implementation. The predicted results of the model were the same in daily QA, while the deviations between the repeated monthly measured GPRs were all within 2%. CONCLUSIONS: The performance of the ACLR model in multi-institution scenarios was validated on a large scale. Routine QA of the ACLR model was established and the model could be used for VMAT PSQA clinically.

Commissioning of optical surface imaging systems for cranial frameless stereotactic radiosurgery.

PURPOSE: This study aimed to evaluate and compare different system calibration methods from a large cohort of systems to establish a commissioning procedure for surface-guided frameless cranial stereotactic radiosurgery (SRS) with intrafractional motion monitoring and gating. Using optical surface imaging (OSI) to guide non-coplanar SRS treatments, the determination of OSI couch-angle dependency, baseline drift, and gated-delivered-dose equivalency are essential. METHODS: Eleven trained physicists evaluated 17 OSI systems at nine clinical centers within our institution. Three calibration methods were examined, including 1-level (2D), 2-level plate (3D) calibration for both surface image reconstruction and isocenter determination, and cube phantom calibration to assess OSI-megavoltage (MV) isocenter concordance. After each calibration, a couch-angle dependency error was measured as the maximum registration error within the couch rotation range. A head phantom was immobilized on the treatment couch and the isocenter was set in the middle of the brain, marked with the room lasers. An on-site reference image was acquired at couch zero, the facial region of interest (ROI) was defined, and static verification images were captured every 10° for 0°-90° and 360°-270°. The baseline drift was assessed with real-time monitoring of the motionless phantom over 20 min. The gated-delivered-dose equivalency was assessed using the electron portal imaging device and gamma test (1%/1mm) in reference to non-gated delivery. RESULTS: The maximum couch-angle dependency error occurs in longitudinal and lateral directions and is reduced significantly (P < 0.05) from 1-level (1.3 ± 0.4 mm) to 2-level (0.8 ± 0.3 mm) calibration. The MV cube calibration does not further reduce the couch-angle dependency error (0.8 ± 0.2 mm) on average. The baseline drift error plateaus at 0.3 ± 0.1 mm after 10 min. The gated-delivered-dose equivalency has a >98% gamma-test passing rate. CONCLUSION: A commissioning method is recommended using the 3D plate calibration, which is verified by radiation isocenter and validated with couch-angle dependency, baseline drift, and gated-delivered-dose equivalency tests. This method characterizes OSI uncertainties, ensuring motion-monitoring accuracy for SRS treatments.

Multi-task autoencoder based classification-regression model for patient-specific VMAT QA.

Patient-specific quality assurance (PSQA) of volumetric modulated arc therapy (VMAT) to assure accurate treatment delivery is resource-intensive and time-consuming. Recently, machine learning has been increasingly investigated in PSQA results prediction. However, the classification performance of models at different criteria needs further improvement and clinical validation (CV), especially for predicting plans with low gamma passing rates (GPRs). In this study, we developed and validated a novel multi-task model called autoencoder based classification-regression (ACLR) for VMAT PSQA. The classification and regression were integrated into one model, both parts were trained alternatively while minimizing a defined loss function. The classification was used as an intermediate result to improve the regression accuracy. Different tasks of GPRs prediction and classification based on different criteria were trained simultaneously. Balanced sampling techniques were used to improve the prediction accuracy and classification sensitivity for the unbalanced VMAT plans. Fifty-four metrics were selected as inputs to describe the plan modulation-complexity and delivery-characteristics, while the outputs were PSQA GPRs. A total of 426 clinically delivered VMAT plans were used for technical validation (TV), and another 150 VMAT plans were used for CV to evaluate the generalization performance of the model. The ACLR performance was compared with the Poisson Lasso (PL) model and found significant improvement in prediction accuracy. In TV, the absolute prediction error (APE) of ACLR was 1.76%, 2.60%, and 4.66% at 3%/3 mm, 3%/2 mm, and 2%/2 mm, respectively; whereas the APE of PL was 2.10%, 3.04%, and 5.29% at 3%/3 mm, 3%/2 mm, and 2%/2 mm, respectively. No significant difference was found between CV and TV in prediction accuracy. ACLR model set with 3%/3 mm can achieve 100% sensitivity and 83% specificity. The ACLR model could classify the unbalanced VMAT QA results accurately, and it can be readily applied in clinical practice for virtual VMAT QA.

Report of AAPM Task Group 235 Radiochromic Film Dosimetry: An Update to TG-55.

The use of radiochromic film (RCF) dosimetry in radiation therapy is extensive due to its high level of achievable accuracy for a wide range of dose values and its suitability under a variety of measurement conditions. However, since the publication of the 1998 AAPM Task Group 55, Report No. 63 on RCF dosimetry, the chemistry, composition, and readout systems for RCFs have evolved steadily. There are several challenges in using the new RCFs, readout systems and validation of the results depending on their applications. Accurate RCF dosimetry requires understanding of RCF selection, handling and calibration methods, calibration curves, dose conversion methods, correction methodologies as well as selection, operation and quality assurance (QA) programs of the readout systems. Acquiring this level of knowledge is not straight forward, even for some experienced users. This Task Group report addresses these issues and provides a basic understanding of available RCF models, dosimetric characteristics and properties, advantages and limitations, configurations, and overall elemental compositions of the RCFs that have changed over the past 20 yr. In addition, this report provides specific guidelines for data processing and analysis schemes and correction methodologies for clinical applications in radiation therapy.

Integration of AI and Machine Learning in Radiotherapy QA.

The use of machine learning and other sophisticated models to aid in prediction and decision making has become widely popular across a breadth of disciplines. Within the greater diagnostic radiology, radiation oncology, and medical physics communities promising work is being performed in tissue classification and cancer staging, outcome prediction, automated segmentation, treatment planning, and quality assurance as well as other areas. In this article, machine learning approaches are explored, highlighting specific applications in machine and patient-specific quality assurance (QA). Machine learning can analyze multiple elements of a delivery system on its performance over time including the multileaf collimator (MLC), imaging system, mechanical and dosimetric parameters. Virtual Intensity-Modulated Radiation Therapy (IMRT) QA can predict passing rates using different measurement techniques, different treatment planning systems, and different treatment delivery machines across multiple institutions. Prediction of QA passing rates and other metrics can have profound implications on the current IMRT process. Here we cover general concepts of machine learning in dosimetry and various methods used in virtual IMRT QA, as well as their clinical applications.

Commissioning and Evaluation of a Third-Party 6 Degrees-of-Freedom Couch Used in Radiotherapy.

PURPOSES: The newly released Protura 6 degrees-of-freedom couch (CIVCO) has limited quality assurance protocols and pertinent publications. Herein, we report our experiences of the Protura system acceptance, commissioning, and quality assurance. METHODS: The Protura system integration was tested with peripheral equipment on the following items: couch movement range limit, 6 degrees-of-freedom movement accuracy, weight test and couch sagging, system connection with Linac, isocentricity of couch and rotation alignment, kV and cone-beam computed tomography imaging of HexaCHECK with MIMI phantom (Standard Imaging), and an in-house custom 6 degrees-of-freedom quality assurance phantom. A couch transmission measurement was also performed. RESULTS: The vertical, longitudinal, and lateral ranges of the 6 degrees-of-freedom couch pedestal are 43.9 to 0.0 cm, 24.6 to 149.5 cm, -20.6 to 20.7 cm, respectively. The couch movement accuracy was within 1 mm in all directions. The couch sagging with a 200 lbs (∼91 kg) evenly distributed object is 1.0 cm and 0.4° pitch in the distal end of the couch. The isocentricity of the couch was about 0.5 mm in diameter of all crosshair projections on the couch isocenter level, and the largest couch rotation alignment observed was (0.3°) at the couch angle of 90°. The deviation from the reference position (zero position) of the HexaCHECK phantom, measured by matching the cone-beam computed tomography with the reference planning computed tomography, was found to be below 0.2 mm in the anterior-posterior and right-left dimensions, 0.4 mm in superior-inferior dimension, and 0.1° in roll, pitch, and yaw directions. CONCLUSIONS: A 6 degrees-of-freedom quality assurance phantom is helpful for the commissioning and routine quality assurance tests. Due to the third-party integration with Linac, the system is prone to "double-correction" errors. A rigorous quality assurance program is the key to a successful clinical implementation of the Protura system.

Machine Learning for Patient-Specific Quality Assurance of VMAT: Prediction and Classification Accuracy.

PURPOSE: To assess the accuracy of machine learning to predict and classify quality assurance (QA) results for volumetric modulated arc therapy (VMAT) plans. METHODS AND MATERIALS: Three hundred three VMAT plans, including 176 gynecologic cancer and 127 head and neck cancer plans, were chosen in this study. Fifty-four complexity metrics were extracted from the QA plans and considered as inputs. Patient-specific QA was performed, and gamma passing rates (GPRs) were used as outputs. One Poisson lasso (PL) regression model was developed, aiming to predict individual GPR, and 1 random forest (RF) classification model was developed to classify QA results as "pass" or "fail." Both technical validation (TV) and clinical validation (CV) were used to evaluate the model reliability. GPR prediction accuracy of PL and classification performance of PL and RF were evaluated. RESULTS: In TV, the mean prediction error of PL was 1.81%, 2.39%, and 4.18% at 3%/3 mm, 3%/2 mm, and 2%/2 mm, respectively. No significant differences in prediction errors between TV and CV were observed. In QA results classification, PL had a higher specificity (accurately identifying plans that can pass QA), whereas RF had a higher sensitivity (accurately identifying plans that may fail QA). By using 90% as the action limit at a 3%/2 mm criterion, the specificity of PL and RF was 97.5% and 87.7% in TV and 100% and 71.4% in CV, respectively. The sensitivity of PL and RF was 31.6% and 100% in TV and 33.3% and 100% in CV, respectively. With 100% sensitivity, the QA workload of 81.2% of plans in TV and 62.5% of plans in CV could be reduced by RF. CONCLUSIONS: The PL model could accurately predict GPR for most VMAT plans. The RF model with 100% sensitivity was preferred for QA results classification. Machine learning can be a useful tool to assist VMAT QA and reduce QA workload.

Impact of delivery characteristics on dose delivery accuracy of volumetric modulated arc therapy for different treatment sites.

This study aimed to investigate the impact of delivery characteristics on the dose delivery accuracy of volumetric modulated arc therapy (VMAT) for different treatment sites. The pretreatment quality assurance (QA) results of 344 VMAT patients diagnosed with gynecological (GYN), head and neck (H&N), rectal or prostate cancer were randomly chosen in this study. Ten metrics reflecting VMAT delivery characteristics were extracted from the QA plans. Compared with GYN and rectal plans, H&N and prostate plans had higher aperture complexity and monitor units (MU), and smaller aperture area. Prostate plans had the smallest aperture area and lowest leaf speed compared with other plans (P < 0.001). No differences in gantry speed were found among the four sites. The gamma passing rates (GPRs) of GYN, rectal and H&N plans were inversely associated with union aperture area (UAA) and leaf speed (Pearson's r: -0.39 to -0.68). GPRs of prostate plans were inversely correlated with aperture complexity, MU and small aperture score (SAS) (absolute Pearson's r: 0.34 to 0.49). Significant differences in GPR between high SAS and low SAS subgroups were found only when leaf speed was <0.42 cm s-1 (P < 0.001). No association of GPR with gantry speed was found in four sites. Leaf speed was more strongly associated with UAA. Aperture complexity and MU were more strongly associated with SAS. VMAT plans from different sites have distinct delivery characteristics. Affecting dose delivery accuracy, leaf speed is the key factor for GYN, rectal and H&N plans, while aperture complexity, MU and small apertures have a higher influence on prostate plans.

Towards real-time respiratory motion prediction based on long short-term memory neural networks.

Radiation therapy of thoracic and abdominal tumors requires incorporating the respiratory motion into treatments. To precisely account for the patient's respiratory motions and predict the respiratory signals, a generalized model for predictions of different types of patients' respiratory motions is desired. The aim of this study is to explore the feasibility of developing a long short-term memory (LSTM)-based generalized model for the respiratory signal prediction. To achieve that, 1703 sets of real-time position management (RPM) data were collected from retrospective studies across three clinical institutions. These datasets were separated as the training, internal validity and external validity groups. Among all the datasets, 1187 datasets were used for model development and the remaining 516 datasets were used to test the model's generality power. Furthermore, an exhaustive grid search was implemented to find the optimal hyper-parameters of the LSTM model. The hyper-parameters are the number of LSTM layers, the number of hidden units, the optimizer, the learning rate, the number of epochs, and the length of time lags. The obtained model achieved superior accuracy over conventional artificial neural network (ANN) models: with the prediction window equaling to 500 ms, the LSTM model achieved an average relative mean absolute error (MAE) of 0.037, an average root mean square error (RMSE) of 0.048, and a maximum error (ME) of 1.687 in the internal validity data, and an average relative MAE of 0.112, an average RMSE of 0.139 and an ME of 1.811 in the external validity data. Compared to the LSTM model trained with default hyper-parameters, the MAE of the optimized model results decreased by 20%, indicating the importance of tuning the hyper-parameters of LSTM models to obtain superior accuracies. This study demonstrates the potential of deep LSTM models for the respiratory signal prediction and illustrates the impacts of major hyper-parameters in LSTM models.