Explore the vast range of titles available.

Purpose The nonwater‐equivalent energy response of electronic portal imaging devices (EPIDs) is a major obstacle to using them for linear accelerator (linac) beam parameter verification. In this study, we propose an EPID‐based machine quality assurance (QA) application that uses a model‐based radiation transport algorithm to convert EPID‐measured images into water‐equivalent dose distributions that can be used to assess beam flatness and symmetry. Methods An in‐house developed, model‐based radiation transport algorithm was used to estimate the incident beam fluence from measured EPID images and convert it into either 3D dose distributions in a virtual water tank or 2D water‐equivalent dose distributions in a virtual ion chamber array. The conversion algorithm was validated using independent measurements in a scanning water tank and a reference ion chamber array under symmetric and also intentionally detuned (i.e., asymmetric) beam conditions. Results For symmetric fields, EPID‐reconstructed percentage depth dose distributions (PDDs) agreed with water tank measurements to within 1% beyond the first 10 mm of depth. Beam profile comparisons showed differences within 1% in low dose‐gradient regions. For all symmetric and intentionally asymmetric fields, beam flatness and symmetry derived from reconstructed images agreed with reference measurements to within 0.2% and 0.3%, respectively. The model demonstrated high sensitivity to the controlled beam asymmetries and steering distortions, with EPID‐reconstructed metrics closely matching reference water‐equivalent measurements and significantly outperforming metrics derived from raw EPID images. Conclusions The proposed model‐based algorithm enables accurate conversion of EPID images into water‐equivalent dose distributions, facilitating accurate determination of beam flatness and symmetry. This application addresses some limitations of the previously proposed EPID‐based linac QA techniques, which are limited to nonwater‐equivalent constancy checks, and supports the use of EPIDs as robust dosimetry tools for linac radiation beam parameter verification.

Purpose Recently, dosimetri applications of the electronic portal imaging device (EPID) in radiotherapy have gained popularity. Confidence in the robust and reliable dosimetric performance of EPID detectors is essential for their clinical use. This study aimed to evaluate the dosimetric performance of the a‐Si 1200 EPID and assess the long‐term stability of its response. Methods Weekly measurements were performed on two clinically used TrueBeam linear accelerators (linacs) equipped with a‐Si 1200 EPID detectors over a 2‐year period. They included dark and flood calibration fields, and EPID response to an open field corrected for the long‐term machine output drift measured with the secondary absolute dosimeters: an ion chamber and an ion chamber array. All measurements were performed using five photon beam energies and two imaging modes: continuous and dosimetry. The measurements were analyzed for constancy and the presence of long‐term trends. Comparisons were made between the two linacs for each beam energy. Pixel sensitivity matrices (PSM) were determined semi‐annually and analyzed for long‐term constancy for both treatment machines. Results The long‐term variation of the dark and flood field signals, integrated across the EPID plane, over the entire observation period did not exceed 0.17% and 0.79%, respectively. The output‐corrected EPID response showed long‐term variation from 0.28% to 0.36%, depending on beam energy, while the short‐term variation was 0.04%–0.07% for EPID and 0.02%–0.06% for secondary dosimeters. The long‐term variation of secondary dosimeters was 0.2%–0.3%. PSMs were found to be stable to within 1% for 97.8% of pixels and 2% for 100% of pixels. Conclusion Techniques to monitor and assess the long‐term performance of the a‐Si 1200 EPID as a dosimeter were developed and implemented using two TrueBeam linacs. The long‐term variation of the EPID response was within clinical tolerance indicated in AAPM TG‐142 report, and the detector was shown to be stable and reproducible for routine clinical dosimetry.

Purpose To develop an alternative computational approach for EPID‐based non‐transit dosimetry using a convolutional neural network model. Method A U‐net followed by a non‐trainable layer named True Dose Modulation recovering the spatialized information was developed. The model was trained on 186 Intensity‐Modulated Radiation Therapy Step & Shot beams from 36 treatment plans of different tumor locations to convert grayscale portal images into planar absolute dose distributions. Input data were acquired from an amorphous‐Silicon Electronic Portal Image Device and a 6 MV X‐ray beam. Ground truths were computed from a conventional kernel‐based dose algorithm. The model was trained by a two‐step learning process and validated through a five‐fold cross‐validation procedure with sets of training and validation of 80% and 20%, respectively. A study regarding the dependance of the amount of training data was conducted. The performance of the model was evaluated from a quantitative analysis based the ϒ‐index, absolute and relative errors computed between the inferred dose distributions and ground truths for six square and 29 clinical beams from seven treatment plans. These results were also compared to those of an existing portal image‐to‐dose conversion algorithm. Results For the clinical beams, averages of ϒ‐index and ϒ‐passing rate (2%‐2mm > 10% Dmax) of 0.24 (±0.04) and 99.29 (±0.70)% were obtained. For the same metrics and criteria, averages of 0.31 (±0.16) and 98.83 (±2.40)% were obtained with the six square beams. Overall, the developed model performed better than the existing analytical method. The study also showed that sufficient model accuracy can be achieved with the amount of training samples used. Conclusion A deep learning‐based model was developed to convert portal images into absolute dose distributions. The accuracy obtained shows that this method has great potential for EPID‐based non‐transit dosimetry.

The gamma analysis metric is a commonly used metric for VMAT plan evaluation. The major drawback of this is the lack of correlation between gamma passing rates and DVH values. The novel GDSAmean metric was developed by Steers et al. to quantify changes in the PTV mean dose (Dmean) for VMAT patients. The aim of this work is to apply the GDSA retrospectively on head‐and‐neck cancer patients treated on the newly acquired Varian Halcyon, to assess changes in GDSAmean, and to evaluate the cause of day‐to‐day changes in the time‐plot series. In‐vivo EPID transmission images of head‐and‐neck cancer patients treated between August 2019 and July 2020 were analyzed retrospectively. The GDSAmean was determined for all patients treated. The changes in patient anatomy and rotational errors were quantified using the daily CBCT images and added to a time‐plot with the daily change in GDSAmean. Over 97% of the delivered treatment fractions had a GDSAmean < 3%. Thirteen of the patients received at least one treatment fraction where the GDSAmean > 3%. Most of these deviations occurred for the later fractions of radiotherapy treatment. Additionally, 92% of these patients were treated for malignancies involving the larynx and oropharynx. Notable deviations in the effective separation diameters were observed for 62% of the patients where the change in GDSAmean > 3%. For the other five cases with GDSAmean < 3%, the mean pitch, roll, and yaw rotational errors were 0.90°, 0.45°, and 0.43°, respectively. A GDSAmean > 3% was more likely due to a change in separation, whereas a GDSAmean < 3% was likely caused by rotational errors. Pitch errors were shown to be the most dominant. The GDSAmean is easily implementable and can aid in scheduling new CT scans for patients before significant deviations in dose delivery occur.

We developed a transformer-based multimodal neural network to predict the gamma passing rate (GPR) in stereotactic body radiation therapy (SBRT) patient-specific quality assurance. Using 1265 SBRT beams from two institutions, the model incorporated portal dose prediction fluence maps with beam complexity descriptors such as modulation complexity score and monitor units. A multi-scale visual-textual transformer, integrating a ViT encoder and feedforward network through a fusion head, was compared with state-of-the-art CNNs across nine gamma criteria. Our approach consistently achieved the lowest root mean squared error (RMSE) and mean absolute error (MAE), with values ranging from 0.785% to 4.258% and 0.418% to 3.197%, respectively, and ablation studies highlighted the necessity of multimodal fusion and multi-scale design. These results demonstrate superior predictive accuracy and generalizability, underscoring the potential of transformer-based multimodal learning to enhance treatment optimization and clinical QA efficiency.

Abstract This study evaluated the impact of off-center diagonal (OCD) profile depth pairing between the treatment planning system (TPS) and the electronic portal imaging device (EPID) on gamma pass rates in portal dosimetry. In clinical workflows, OCD profiles are used in the TPS to generate predicted images via the portal dosimetry image prediction (PDIP) algorithm and in the EPID system to correct measured fluence. The consistency of these settings may influence verification accuracy. Portal images were acquired using a TrueBeam linear accelerator with an aS1200 EPID for four photon energies: 6X, 10X, 6 flattening filter-free (FFF) and 10FFF. Five OCD profiles (reference depth, 5, 10, 20 and 30 cm) were configured in both the PDIP model and EPID system. For each energy, a total of 175 plan–measurement combinations were evaluated, derived from five PDIP OCD depths combined with five EPID OCD depths across seven field sizes. Field sizes ranged from 5 × 5 to 30 × 30 cm2. Gamma analysis used 3%/3 mm criteria with a 10% dose threshold. A two-way analysis of variance assessed the effects of TPS and EPID OCD depths and their interaction. For 6X and 10X beams, pass rates varied with configuration, showing better agreement when depths were matched or EPID was deeper. In contrast, 6FFF and 10FFF beams maintained high pass rates with minimal variation. These findings indicate that OCD depth pairing influences portal dosimetry performance, particularly for flattened beams, underscoring the importance of depth-aware configuration in QA protocols.

Purpose Multi‐leaf‐collimator (MLC) leaf position accuracy is important for accurate dynamic radiotherapy treatment plan delivery. Machine log files have become widely utilized for quality assurance (QA) of such dynamic treatments. The primary aim is to test the sensitivity of machine log files in comparison to electronic portal imaging device (EPID)‐based measurements to MLC position errors caused by leaf backlash. The secondary aim is to investigate the effect of MLC leaf backlash on MLC leaf motion during clinical dynamic plan delivery. Methods The sensitivity of machine log files and two EPID‐based measurements were assessed via a controlled experiment, whereby the length of the “T” section of a series of 12 MLC leaf T‐nuts in a Varian Millennium MLC for a Trilogy C‐series type linac was reduced by sandpapering the top of the “T” to introduce backlash. The built‐in machine MLC leaf backlash test as well as measurements for two EPID‐based dynamic MLC positional tests along with log files were recorded pre‐ and post‐T‐nut modification. All methods were investigated for sensitivity to the T‐nut change by assessing the effect on measured MLC leaf positions. A reduced version of the experiment was repeated on a TrueBeam type linac with Millennium MLC. Results No significant differences before and after T‐nut modification were detected in any of the log file data. Both EPID methods demonstrated sensitivity to the introduced change at approximately the expected magnitude with a strong dependence observed with gantry angle. EPID‐based data showed MLC positional error in agreement with the micrometer measured T‐nut length change to 0.07 ± 0.05 mm (1 SD) using the departmental routine QA test. Backlash results were consistent between linac types. Conclusion Machine log files appear insensitive to MLC position errors caused by MLC leaf backlash introduced via the T‐nut. The effect of backlash on clinical MLC motions is heavily gantry angle dependent.

Considering the contradiction among the response speed, overshoot and stability of system when the motor load system adopts PID control, a control strategy combining RBF (Radial Basis Function) neural network supervisory control and expert PID control is designed to effectively improve this problem in this paper. First of all, the related algorithms of RBF neural network supervisory control composed of RBF neural network and PID control (RSC-PID) is introduced. This method can make the motor load system reach a steady state faster than simple PID control. But RSC-PID is also unsatisfactory in terms of overshoot. Based on the RSC-PID control method, a hybrid controller combining RBF neural network supervisory control and expert PID control (RSC-EPID) is proposed. This method combines RSC-PID control theory with expert PID control ideas, further improves the stability and rapidity of the system, reduces the overshoot. Moreover, when the input signal is a time-varying signal with interference, the motor load system shows better anti-interference performance after using RSC-EPID. The simulation results show that RSC-EPID control improves the tracking effect of the output signal of the motor load system, ensures the stability of the system, and improves the performance of the system.

Background Independent verification of the dose delivered by complex radiotherapy can be performed by electronic portal imaging device (EPID) dosimetry. This paper presents 5‐yr EPID in vivo dosimetry (IVD) data obtained using the Dosimetry Check (DC) software on a large cohort including breast, lung, prostate, and head and neck (H&N) cancer patients. Material and Methods The difference between in vivo dose measurements obtained by DC and point doses calculated by the Eclipse treatment planning system was obtained on 3795 radiotherapy patients treated with volumetric modulated arc therapy (VMAT) (n = 842) and three‐dimensional conformal radiotherapy (3DCRT) (n = 2953) at 6, 10, and 15 MV. In cases where the dose difference exceeded ±10% further inspection and additional phantom measurements were performed. Results The mean and standard deviation (μ±σ) of the percentage difference in dose obtained by DC and calculated by Eclipse in VMAT was: 0.19±3.89% in brain, 1.54±4.87% in H&N, and 1.23±4.61% in prostate cancer. In 3DCRT, this was 1.79±3.51% in brain, −2.95±5.67% in breast, −1.43±4.38% in bladder, 1.66±4.77% in H&N, 2.60 ± 5.35% in lung and −3.62±4.00% in prostate cancer. A total of 153 plans exceeded the ±10% alert criteria, which included: 88 breast plans accounting for 7.9% of all breast treatments; 28 H&N plans accounting for 4.4% of all H&N treatments; and 12 prostate plans accounting for 3.5% of all prostate treatments. All deviations were found to be as a result of patient‐related anatomical deviations and not from procedural errors. Conclusions This preliminary data shows that EPID‐based IVD with DC may not only be useful in detecting errors but has the potential to be used to establish site‐specific dose action levels. The approach is straightforward and has been implemented as a radiographer‐led service with no disruption to the patient and no impact on treatment time.