In vivo EPID-based daily treatment error identification for volumetric-modulated arc therapy in head and neck cancers with a hierarchical convolutional neural network: a feasibility study

We proposed a deep learning approach to classify various error types in daily VMAT treatment of head and neck cancer patients based on EPID dosimetry, which could provide additional information to support clinical decisions for adaptive planning. 146 arcs from 42 head and neck patients were analyzed. Anatomical changes and setup errors were simulated in 17,820 EPID images of 99 arcs obtained from 30 patients using in-house software for model training, validation, and testing. Subsequently, 141 clinical EPID images from 47 arcs belonging to the remaining 12 patients were utilized for clinical testing. The hierarchical convolutional neural network (HCNN) model was trained to classify error types and magnitudes using EPID dose difference maps. Gamma analysis with 3%/2 mm (dose difference/distance to agreement) criteria was also performed. The F1 score, a combination of precision and recall, was utilized to evaluate the performance of the HCNN model and gamma analysis. The adaptive fractioned doses were calculated to verify the HCNN classification results. For error type identification, the overall F1 score of the HCNN model was 0.99 and 0.91 for primary type and subtype identification, respectively. For error magnitude identification, the overall F1 score in the simulation dataset was 0.96 and 0.70 for the HCNN model and gamma analysis, respectively; while the overall F1 score in the clinical dataset was 0.79 and 0.20 for the HCNN model and gamma analysis, respectively. The HCNN model-based EPID dosimetry can identify changes in patient transmission doses and distinguish the treatment error category, which could potentially provide information for head and neck cancer treatment adaption.

A feasibility study for in vivo treatment verification of IMRT using Monte Carlo dose calculation and deep learning-based modelling of EPID detector response

Background

This paper describes the development of a predicted electronic portal imaging device (EPID) transmission image (TI) using Monte Carlo (MC) and deep learning (DL). The measured and predicted TI were compared for two-dimensional in vivo radiotherapy treatment verification.

Methods

The plan CT was pre-processed and combined with solid water and then imported into PRIMO. The MC method was used to calculate the dose distribution of the combined CT. The U-net neural network-based deep learning model was trained to predict EPID TI based on the dose distribution of solid water calculated by PRIMO. The predicted TI was compared with the measured TI for two-dimensional in vivo treatment verification.

Results

The EPID TI of 1500 IMRT fields were acquired, among which 1200, 150, and 150 fields were used as the training set, the validation set, and the test set, respectively. A comparison of the predicted and measured TI was carried out using global gamma analyses of 3%/3 mm and 2%/2 mm (5% threshold) to validate the model's accuracy. The gamma pass rates were greater than 96.7% and 92.3%, and the mean gamma values were 0.21 and 0.32, respectively.

Conclusions

Our method facilitates the modelling process more easily and increases the calculation accuracy when using the MC algorithm to simulate the EPID response, and has potential to be used for in vivo treatment verification in the clinic.

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.

Simple Electronic Portal Imager-Based Pretreatment Quality Assurance using Acuros XB: A Feasibility Study

Objective

This study demonstrates a novel electronic portal imaging device (EPID)-based forward dosimetry approach for pretreatment quality assurance aided by a treatment planning system (TPS).

Materials and Methods

Dynamic multileaf collimator intensity-modulated radiation therapy (IMRT) plans were delivered in EPID and fluence was captured on a beam-by-beam basis (F_EPID). An open field having dimensions equal to those of the largest IMRT field was used in the TPS to obtain the transmitted fluence. This represented the patient-specific heterogeneity correction (F_het). To obtain the resultant heterogeneity-corrected fluence, EPID measured fluence was corrected for the TPS generated heterogeneity (F_Resultant = F_EPID × F_het). Next, the calculated fluence per beam basis was collected from TPS (F_TPS). Finally, F_Resultant and F_TPS were compared using a 3% percentage dose difference (% DD)-3 mm distance to agreement [DTA] gamma analysis in an isocentric plane (two-dimensional [2D]) and multiple planes (quasi three-dimensional [3D]) orthogonal to the beam axis.

Results

The 2D heterogeneity-corrected dose reconstruction revealed a mean γ passing of the pelvis, thorax, and head-and-neck cases of 96.3% ±2.0%, 96.3% ±1.8%, and 96.1% ±2.2%, respectively. Quasi-3D γ passing for the pelvis, thorax, and head-and-neck cases were 97.5% ±1.4%, 96.3% ±2.4%, and 97.5% ±1.0%, respectively.

Conclusion

EPID dosimetry produced an inferior result due to no heterogeneity corrections for sites such as the lung and esophagus. Incorporating TPS-based heterogeneity correction improved the EPID dosimetry result for those sites with large heterogeneity.

Electronic Portal Imaging Device-Based Three-Dimensional Volumetric Dosimetry for Intensity-modulated Radiotherapy Pretreatment Quality Assurance

Aim:

This study aimed at evaluating the efficacy of treatment planning system (TPS)-based heterogeneity correction for two- and three-dimensional (2D and 3D) electronic portal imaging device (EPID)-based pretreatment dose verification. An experiment was conducted on the EPID back-projection technique and intensity-modulated radiotherapy (IMRT).

Materials and Methods:

Treatment plans were delivered in EPID without a patient to obtain the fluence pattern (F_EPID). A heterogeneity correction plane (F_het) for an open beam of 30 cm × 30 cm was extracted from the TPS. The heterogeneity-corrected measured fluence is developed by matrix element multiplication (F_Resultant = F_EPID × F_het). Further planes were summed to develop a 3D dose distribution and exported to the TPS. Dose verifications for 2D and 3D were carried out with the corresponding TPS values using 2D gamma analysis (ɣ) and dose volume histogram (DVH) comparison, respectively. Totally, 33 patients (17 head–neck and 16 thorax cases) were evaluated in this study.

Results:

The head–neck and thorax plans show a 3-mm-distance to agreement (DTA) 3% DD gamma passing of 96.3% ± 2.0% and 95.4% ± 1.8% points, respectively, between F_TPS and F_Resultant. The comparison of the uncorrected measured fluence (F_EPID) with F_TPS reveals a gamma passing of 82.2% ± 7.3% and 80.4% ± 8.6% for head–neck and thorax cases, respectively. A total of 87 out of the 102 head–neck and thorax beams exhibit a planner gamma passing of 97.6% ± 2.1%. In the 3D-DVH comparison of thorax and head–neck cases, D5% for planning target volume were −0.5% ± 2.2% and −2.1% ± 3.5%, respectively; D95% varies as 1.0% ± 2.7% and 1.4% ± 1.1% between TPS calculated and heterogeneity-corrected-EPID-based dose reconstruction.

Conclusion:

The novel TPS-based heterogeneity correction can improve the 2D and 3D EPID-based back projection technique. Structures with large heterogeneities can also be handled using the proposed technique.

First Report of the Clinical Use of a Commercial Automated System for Daily Patient QA Using EPID Exit Images

Purpose

To characterize the clinical utility of a new commercially available system for daily patient treatment quality assurance using electronic portal imaging detector (EPID) exit dose images.

Methods and Materials

The PerFRACTION automated quality assurance system was used to acquire integrated EPID images for every field every day for 60 treatment courses for 57 patients. Four thousand seventy-nine field values from 855 fractions were analyzed. Gamma passing rates were computed by the system for each field daily. Passing rates and pass-fail status were recorded by treatment modality (intensity modulated radiation therapy or 3-dimensional conformal radiotherapy) and location. When failures occurred, an attempt was made to determine the reason.

Results

Overall, 23% and 8% of fields failed at 2%/2 mm and 3%/3 mm, respectively. Forty-eight percent and 24% of fields failed at least once during the course of therapy for the 2 tolerance settings. Eighteen percent and 8% of all fractions failed and 60% and 28% of courses failed for the 2 tolerance settings, respectively. Eighteen percent of daily field passing rates were below 75% for 3%/3 mm tolerances. Intensity modulated radiation therapy had higher passing rates than 3-dimensional conformal radiation therapy. For 3%/3 mm tolerances, the fraction fail rate for the brain, extremity, and spine treatment sites failed the least, whereas the abdomen, chest, and head and neck failed more often. The most commonly identified reason for failure was body position change, but the reason for about half the daily field value failures could not be identified.

Conclusions

This is the first report of the clinical utility of a commercial daily patient treatment quality assurance system using EPID exit images. Variations were found in a clinically relevant percentage of images, and these potentially indicate important treatment variations. Reasons for failures are not always discernable. The system was practical to use because of automation and continues to be used for monitoring of nearly every patient in every field every day.

EPID-based in vivo dosimetry using Dosimetry Check™: Overview and clinical experience in a 5-yr study including breast, lung, prostate, and head and neck cancer patients

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.

In-vivo EPID dosimetry for IMRT and VMAT based on through-air predicted portal dose algorithm

We have adapted the methodology of Berry et al. (2012) for Intensity Modulated Radiotherapy (IMRT) and Volumetric Modulated Arc Therapy (VMAT) treatments at a fixed source to imager distance (SID) based on the manufacturer's through-air portal dose image prediction algorithm. In order to fix the SID a correction factor was introduced to account for the change in air gap between patient and imager. Commissioning data, collected with multiple field sizes, solid water thicknesses and air gaps, were acquired at 150 cm SID on the Varian aS1200 EPID. The method was verified using six IMRT and seven VMAT plans on up to three different phantoms. The method's sensitivity and accuracy were investigated by introducing errors. A global 3%/3 mm gamma was used to assess the differences between the predicted and measured portal dose images. The effect of a varying air gap on EPID signal was found to be significant - varying by up to 30% with field size, phantom thickness, and air gap. All IMRT plans passed the 3%/3 mm gamma criteria by more than 95% on the three phantoms. 23 of 24 arcs from the VMAT plans passed the 3%/3 mm gamma criteria by more than 95%. This method was found to be sensitive to a range of potential errors. The presented approach provides fast and accurate in-vivo EPID dosimetry for IMRT and VMAT treatments and can potentially replace many pre-treatment verifications.

Comparison of forward- and back-projection in vivo EPID dosimetry for VMAT treatment of the prostate

In the forward-projection method of portal dosimetry for volumetric modulated arc therapy (VMAT), the integrated signal at the electronic portal imaging device (EPID) is predicted at the time of treatment planning, against which the measured integrated image is compared. In the back-projection method, the measured signal at each gantry angle is back-projected through the patient CT scan to give a measure of total dose to the patient. This study aims to investigate the practical agreement between the two types of EPID dosimetry for prostate radiotherapy. The AutoBeam treatment planning system produced VMAT plans together with corresponding predicted portal images, and a total of 46 sets of gantry-resolved portal images were acquired in 13 patients using an iViewGT portal imager. For the forward-projection method, each acquisition of gantry-resolved images was combined into a single integrated image and compared with the predicted image. For the back-projection method, iViewDose was used to calculate the dose distribution in the patient for comparison with the planned dose. A gamma index for 3% and 3 mm was used for both methods. The results were investigated by delivering the same plans to a phantom and repeating some of the deliveries with deliberately introduced errors. The strongest agreement between forward- and back-projection methods is seen in the isocentric intensity/dose difference, with moderate agreement in the mean gamma. The strongest correlation is observed within a given patient, with less correlation between patients, the latter representing the accuracy of prediction of the two methods. The error study shows that each of the two methods has its own distinct sensitivity to errors, but that overall the response is similar. The forward- and back-projection EPID dosimetry methods show moderate agreement in this series of prostate VMAT patients, indicating that both methods can contribute to the verification of dose delivered to the patient.

A dual two dimensional electronic portal imaging device transit dosimetry model based on an empirical quadratic formalism

OBJECTIVE

This study describes a two dimensional electronic portal imaging device (EPID) transit dosimetry model that can predict either: (1) in-phantom exit dose, or (2) EPID transit dose, for treatment verification.

METHODS

The model was based on a quadratic equation that relates the reduction in intensity to the equivalent path length (EPL) of the attenuator. In this study, two sets of quadratic equation coefficients were derived from calibration dose planes measured with EPID and ionization chamber in water under reference conditions. With two sets of coefficients, EPL can be calculated from either EPID or treatment planning system (TPS) dose planes. Consequently, either the in-phantom exit dose or the EPID transit dose can be predicted from the EPL. The model was tested with two open, five wedge and seven sliding window prostate and head and neck intensity-modulated radiation therapy (IMRT) fields on phantoms. Results were analysed using absolute gamma analysis (3%/3 mm).

RESULTS

The open fields gamma pass rates were >96.8% for all comparisons. For wedge and IMRT fields, comparisons between predicted and TPS-computed in-phantom exit dose resulted in mean gamma pass rate of 97.4% (range, 92.3-100%). As for the comparisons between predicted and measured EPID transit dose, the mean gamma pass rate was 97.5% (range, 92.6-100%).

CONCLUSION

An EPID transit dosimetry model that can predict in-phantom exit dose and EPID transit dose was described and proven to be valid.

ADVANCES IN KNOWLEDGE

The described model is practical, generic and flexible to encourage widespread implementation of EPID dosimetry for the improvement of patients' safety in radiotherapy.

Portal dosimetry in wedged beams

Portal dosimetry using electronic portal imaging devices (EPIDs) is often applied to verify high‐energy photon beam treatments. Due to the change in photon energy spectrum, the resulting dose values are, however, not very accurate in the case of wedged beams if the pixel‐to‐dose conversion for the situation without wedge is used. A possible solution would be to consider a wedged beam as another photon beam quality requiring separate beam modeling of the dose calculation algorithm. The aim of this study was to investigate a more practical solution: to make aSi EPID‐based dosimetry models also applicable for wedged beams without an extra commissioning effort of the parameters of the model. For this purpose two energy‐dependent wedge multiplication factors have been introduced to be applied for portal images taken with and without a patient/phantom in the beam. These wedge multiplication factors were derived from EPID and ionization chamber measurements at the EPID level for wedged and nonwedged beams, both with and without a polystyrene slab phantom in the beam. This method was verified for an EPID dosimetry model used for wedged beams at three photon beam energies (6, 10, and 18 MV) by comparing dose values reconstructed in a phantom with data provided by a treatment planning system (TPS), as a function of field size, depth, and off‐axis distance. Generally good agreement, within 2%, was observed for depths between dose maximum and 15 cm. Applying the new model to EPID dose measurements performed during ten breast cancer patient treatments with wedged 6 MV photon beams showed that the average isocenter underdosage of 5.3% was reduced to 0.4%. Gamma‐evaluation (global 3%/3 mm) of these in vivo data showed an increase in percentage of points with γ≤1 from 60.2% to 87.4%, while γmean reduced from 1.01 to 0.55. It can be concluded that, for wedged beams, the multiplication of EPID pixel values with an energy‐dependent correction factor provides good agreement between dose values determined by an EPID and a TPS, indicating the usefulness of such a practical solution.

PACS numbers: 87.55.km, 87.55.kd, 87.55.Qr, 87.56a.ng

Evaluation of the accuracy of dose delivery for IMRT based on transit dosimetry

The authors have developed and evaluated a new method that uses transit dose in the treatment room in verification of dose delivery to patients. Five intensity modulated radiotherapy (IMRT) plans were selected from actual cancer patients, and transit dose was measured using MapCHECK 2 and an anthropomorphic phantom. The criteria used as a verification tool for the dose delivery to the patient were gamma-index-based dose comparison between the computed dose and measured dose. When the doses were delivered to an anthropomorphic phantom normally, the average passing rate was 95.2% based on a gamma index analysis. This feasibility study suggested that transit dose-based quality assurance can provide information about the accuracy of an inhomogeneity correction algorithm and patient positioning during treatment, allowing its use as a verification tool for actual dose delivery to patients in the treatment room.

Time dependent pre-treatment EPID dosimetry for standard and FFF VMAT

Methods to calibrate Megavoltage electronic portal imaging devices (EPIDs) for dosimetry have been previously documented for dynamic treatments such as intensity modulated radiotherapy (IMRT) using flattened beams and typically using integrated fields. While these methods verify the accumulated field shape and dose, the dose rate and differential fields remain unverified. The aim of this work is to provide an accurate calibration model for time dependent pre-treatment dose verification using amorphous silicon (a-Si) EPIDs in volumetric modulated arc therapy (VMAT) for both flattened and flattening filter free (FFF) beams. A general calibration model was created using a Varian TrueBeam accelerator, equipped with an aS1000 EPID, for each photon spectrum 6 MV, 10 MV, 6 MV-FFF, 10 MV-FFF. As planned VMAT treatments use control points (CPs) for optimization, measured images are separated into corresponding time intervals for direct comparison with predictions. The accuracy of the calibration model was determined for a range of treatment conditions. Measured and predicted CP dose images were compared using a time dependent gamma evaluation using criteria (3%, 3 mm, 0.5 sec). Time dependent pre-treatment dose verification is possible without an additional measurement device or phantom, using the on-board EPID. Sufficient data is present in trajectory log files and EPID frame headers to reliably synchronize and resample portal images. For the VMAT plans tested, significantly more deviation is observed when analysed in a time dependent manner for FFF and non-FFF plans than when analysed using only the integrated field. We show EPID-based pre-treatment dose verification can be performed on a CP basis for VMAT plans. This model can measure pre-treatment doses for both flattened and unflattened beams in a time dependent manner which highlights deviations that are missed in integrated field verifications.

Development of multi-planar dose verification by use of a flat panel EPID for intensity-modulated radiation therapy

Our purpose in this study was to evaluate the accuracy of a new multi-planar dose measurement method. The multi-planar dose distributions were reconstructed at each depth by convolution of EPID fluence and dose kernels with the use of EPIDose software (SunNuclear). The EPIDose was compared with EPID, MapCHECK (SunNuclear), EDR2 (Kodak), and Monte Carlo-calculated dose profiles. The EPIDose profiles were almost in agreement with Monte Carlo-calculated dose profiles and MapCHECK for test plans. The dose profiles were in good agreement with EDR2 at the penumbra region. For dose distributions, EPIDose, EDR2, and MapCHECK agreed with that of the treatment-planning system at each depth in the gamma analysis. In comparisons of clinical IMRT plans, EPIDose had almost the same accuracy as MapCHECK and EDR2. Because EPIDose has a wide dynamic range and high resolution, it is a useful tool for the complicated IMRT verification. Furthermore, EPIDose can also evaluate the absolute dose.