Examination of geometric and dosimetric accuracies of gated step-and-shoot intensity modulated radiation therapy

Image-based features in machine learning to identify delivery errors and predict error magnitude for patient-specific IMRT quality assurance

OBJECTIVE

To identify delivery error type and predict associated error magnitude by image-based features using machine learning (ML).

METHODS

In this study, a total of 40 thoracic plans (including 208 beams) were selected, and four error types with different magnitudes were introduced into the original plans, including 1) collimator misalignment (COLL), 2) monitor unit (MU) variation, 3) systematic multileaf collimator misalignment (MLCS), and 4) random MLC misalignment (MLCR). These dose distributions of portal dose predictions for the original plans were defined as the reference dose distributions (RDD), while those for the error-introduced plans were defined as the error-introduced dose distributions (EDD). Both distributions were calculated for all beams with portal dose image prediction (PDIP). Besides, 14 image-based features were extracted from RDD and EDD of portal dose predictions to obtain the feature vectors. In addition, a random forest was adopted for the multiclass classification task, and regression prediction for error magnitude.

RESULTS

The top five features extracted with the highest weight included 1) the relative displacement in the x direction, 2) the ratio of the absolute minimum residual error to the maximal RDD value, 3) the product of the maximum and minimum residuals, 4) the ratio of the absolute maximum residual error to the maximal RDD value, and 5) the ratio of the absolute mean residual value to the maximal RDD value. The relative displacement in the x direction had the highest weight. The overall accuracy of the five-class classification model was 99.85% for the validation set and 99.30% for the testing set. This model could be applied to the classification of the error-free plan, COLL, MU, MLCS, and MLCR with an accuracy of 100%, 98.4%, 99.9%, 98.0%, and 98.3%, respectively. MLCR had the worst performance in error magnitude prediction (70.1-96.6%), while others had better performance in error magnitude prediction (higher than 93%). In the error magnitude prediction, the mean absolute error (MAE) between predicted error magnitude and actual error ranged from 0.03 to 0.33, with the root mean squared error (RMSE) varying from 0.17 to 0.56 for the validation set. The MAE and RMSE ranged from 0.03 to 0.50 and 0.44 to 0.59 for the test set, respectively.

CONCLUSION

It could be demonstrated in this study that the image-based features extracted from RDD and EDD can be employed to identify different types of delivery errors and accurately predict error magnitude with the assistance of ML techniques. They can be used to associate traditional gamma analysis with clinically based analysis for error classification and magnitude prediction in patient-specific IMRT quality assurance.

Delivery efficiency of an Elekta linac under gated operation

In this study, we have characterized the efficiency of an Elekta linac in the delivery of gated radiotherapy. We have explored techniques to reduce the beam‐on delay and to improve the delivery efficiency, and have investigated the impact of frequent beam interruptions on the dosimetric accuracy of gated deliveries. A newly available gating interface was installed on an Elekta Synergy. Gating signals were generated using a surface mapping system in conjunction with a respiratory motion phantom. A series of gated deliveries were performed using volumetric modulated arc therapy (VMAT) treatment plans previously generated for lung cancer patients treated with stereotactic body radiotherapy. Baseline values were determined for the delivery times. The machine was then tuned in an effort to minimize beam‐on delays and improve delivery efficiency. After that process was completed, the dosimetric accuracy of the gated deliveries was evaluated by comparing the measured and the planned coronal dose distributions using gamma index analyses. Comparison of the gated and the non‐gated deliveries were also performed. The results demonstrated that, with the optimal machine settings, the average beam‐on delay was reduced to less than 0.22 s. High dosimetric accuracy was demonstrated with gamma index passing rates no lower than 99.0% for all tests (3%/3 mm criteria). Consequently, Elekta linacs can provide a practical solution for gated VMAT treatments with high dosimetric accuracy and only a moderate increase in the overall delivery time.

PACS numbers: 87.56.bd, 87.55.de, 87.55.ne

Dose verification for respiratory-gated volumetric modulated arc therapy

A novel commercial medical linac system (TrueBeam™, Varian Medical Systems, Palo Alto, CA) allows respiratory-gated volumetric modulated arc therapy (VMAT), a new modality for treating moving tumors with high precision and improved accuracy by allowing for regular motion associated with a patient's breathing during VMAT delivery. The purpose of this work is to adapt a previously-developed dose reconstruction technique to evaluate the fidelity of VMAT treatment during gated delivery under clinic-relevant periodic motion related to patient breathing. A Varian TrueBeam system was used in this study. VMAT plans were created for three patients with lung or pancreas tumors. Conventional 6 and 15 MV beams with flattening filter and high-dose-rate 10 MV beams with no flattening filter were used in these plans. Each patient plan was delivered to a phantom first without gating and then with gating for three simulated respiratory periods (3, 4.5 and 6 s). Using the adapted log-file-based dose reconstruction procedure supplemented with ion chamber array (Seven29™, PTW, Freiburg, Germany) measurements, the delivered dose was used to evaluate the fidelity of gated VMAT delivery. Comparison of Seven29 measurements with and without gating showed good agreement with gamma-index passing rates above 99% for 1%/1 mm dose accuracy/distance-to-agreement criteria. With original plans as reference, gamma-index passing rates were 100% for the reconstituted plans (1%/1 mm criteria) and 93.5-100% for gated Seven29 measurements (3%/3 mm criteria). In the presence of leaf error deliberately introduced into the gated delivery of a pancreas patient plan, both dose reconstruction and Seven29 measurement consistently indicated substantial dosimetric differences from the original plan. In summary, a dose reconstruction procedure was demonstrated for evaluating the accuracy of respiratory-gated VMAT delivery. This technique showed that under clinical operation, the TrueBeam system faithfully realized treatment plans with gated delivery. This methodology affords a useful tool for machine- and patient-specific quality assurance of the newly available respiratory-gated VMAT.

Evaluation of delivered monitor unit accuracy of gated step-and-shoot IMRT using a two-dimensional detector array

PURPOSE

To overcome the problem of organ motion in intensity-modulated radiation therapy (IMRT), gated IMRT is often used for the treatment of lung cancer. In this study, the authors investigated the accuracy of the delivered monitor units (MUs) from each segment during gated IMRT using a two-dimensional detector array for user-specific verification purpose.

METHODS

The authors planned a 6 MV photon, seven-port step-and-shoot lung IMRT delivery. The respiration signals for gated IMRT delivery were obtained from the one-dimensional moving phantom using the real-time position management (RPM) system (Varian Medical Systems, Palo Alto, CA). The beams were delivered using a Clinac iX (Varian Medical Systems, Palo Alto, CA) with the Millennium 120 MLC. The MatriXX (IBA Dosimetry GmbH, Germany) was validated through consistency and reproducibility tests as well as comparison with measurements from a Farmer-type ion chamber. The authors delivered beams with varying dose rates and duty cycles and analyzed the MatriXX data to evaluate MU delivery accuracy.

RESULTS

There was quite good agreement between the planned segment MUs and the MUs computed from the MatriXX within +/- 2% error. The beam-on times computed from the MatriXX data were almost identical for all cases, and they matched well with the RPM beam-on and beam-off signals. A slight difference was observed between them, but it was less than 40 ms. The gated IMRT delivery demonstrated an MU delivery accuracy that was equivalent to ungated IMRT, and the delivered MUs with a gating signal agreed with the planned MUs within +/- 0.5 MU regardless of dose rate and duty cycle.

CONCLUSIONS

The authors can conclude that gated IMRT is able to deliver an accurate dose to a patient during a procedure. The authors believe that the methodology and results can be transferred to other vendors' devices, particularly those that do not provide MLC log data for a verification purpose.

Initial clinical experience with frameless optically guided stereotactic radiosurgery/radiotherapy in pediatric patients

OBJECTIVE

The objective of this study is to report our initial experience treating pediatric patients with central nervous system tumors using a frameless, optically guided linear accelerator.

MATERIALS AND METHODS

Pediatric patients were selected for treatment after evaluation by a multidisciplinary neuro-oncology team including neurosurgery, neurology, pathology, oncology, and radiation oncology. Prior to treatment, all patients underwent treatment planning using magnetic resonance imaging (MRI) and treatment simulation on a standard computed tomography scanner (CT). For CT simulation, patients were fitted with a customized plastic face mask with a bite block attached to an optical array with four reflective markers. After ensuring adequate reproducibility, these markers were tracked during treatment by an infra-red camera. All treatments were delivered on a Varian Trilogy linear accelerator. The follow-up period ranges from 1-18 months, with a median follow-up of 6 months.

RESULTS

Nine patients, ages ranging from 12 to 19 years old (median age 15 years old), with a variety of tumors have been treated. Patients were treated for juvenile pilocytic astrocytoma (JPA; n = 2), pontine low-grade astrocytoma (n = 1), pituitary adenoma (n = 3), metastatic medulloblastoma (n = 1), acoustic neuroma (n = 1), and pineocytoma (n = 1). We followed patients for a median of 12 months (range 3-18 months) with no in-field failures and were able to obtain encouraging toxicity profiles.

CONCLUSION

Frameless stereotactic optically guided radiosurgery and radiotherapy provides a feasible and accurate tool to treat a number of benign and malignant tumors in children with minimal treatment-related morbidity.

The use of EPID-measured leaf sequence files for IMRT dose reconstruction in adaptive radiation therapy

For intensity modulated radiation treatment (IMRT) dose reconstruction, multileaf collimator (MLC) log files have been shown applicable for deriving delivered fluence maps. However, MLC log files are dependent on the accuracy of leaf calibration and only available from one linear accelerator manufacturer. This paper presents a proof of feasibility and principles in (1) using an amorphous silicon electronic portal imaging device (aSi-EPID) to capture the MLC segments during an IMRT delivery and (2) reconstituting a leaf sequence (LS) file based on the leaf end positions calculated from the MLC segments and their associated fractional monitor units. These EPID-measured LS files are then used to derive delivered fluence maps for dose reconstruction. The developed approach was tested on a pelvic phantom treated with a typical prostate IMRT plan. The delivered fluence maps, which were derived from the EPID-measured LS files, showed slight differences in the intensity levels compared with the corresponding planned ones. The dose distribution calculated with the delivered fluence maps showed a discernible difference in the high dose region when compared to that calculated with the planned fluence maps. The maximum dose in the former distribution was also 2.5% less than that in the latter one. The EPID-measured LS file can serve the same purpose as a MLC log files does for the derivation of the delivered fluence map and yet is independent of the leaf calibration. The approach also allows users who do not have access to MLC log files to probe the actual IMRT delivery and translate the information gained for dose reconstruction in adaptive radiation therapy.

Combined kV and MV imaging for real-time tracking of implanted fiducial markers

In the presence of intrafraction organ motion, target localization uncertainty can greatly hamper the advantage of highly conformal dose techniques such as intensity modulated radiation therapy (IMRT). To minimize the adverse dosimetric effect caused by tumor motion, a real-time knowledge of the tumor position is required throughout the beam delivery process. The recent integration of onboard kV diagnostic imaging together with MV electronic portal imaging devices on linear accelerators can allow for real-time three-dimensional (3D) tumor position monitoring during a treatment delivery. The aim of this study is to demonstrate a near real-time 3D internal fiducial tracking system based on the combined use of kV and MV imaging. A commercially available radiotherapy system equipped with both kV and MV imaging systems was used in this work. A hardware video frame grabber was used to capture both kV and MV video streams simultaneously through independent video channels at 30 frames per second. The fiducial locations were extracted from the kV and MV images using a software tool. The geometric tracking capabilities of the system were evaluated using a pelvic phantom with embedded fiducials placed on a moveable stage. The maximum tracking speed of the kV/MV system is approximately 9 Hz, which is primarily limited by the frame rate of the MV imager. The geometric accuracy of the system is found to be on the order of less than 1 mm in all three spatial dimensions. The technique requires minimal hardware modification and is potentially useful for image-guided radiation therapy systems.

Retrospective IMRT dose reconstruction based on cone-beam CT and MLC log-file

PURPOSE

Head-and-neck (HN) cone-beam computed tomography (CBCT) can be exploited to probe the IMRT dose delivered to a patient taking into account the interfraction anatomic variation and any potential inaccuracy in the IMRT delivery. The aim of this work is to reconstruct the intensity-modulated radiation therapy dose delivered to an HN patient using the CBCT and multileaf collimator (MLC) log-files.

METHODS AND MATERIALS

A cylindrical CT phantom was used for calibrating the electron density and validating the procedures of the dose reconstruction. Five HN patients were chosen, and for each patient, CBCTs were performed on three separate fractions spaced every 2 weeks starting from the first fraction. The respective MLC log-files were retrieved and converted into fluence maps. The dose was then reconstructed on the corresponding CBCT with the regenerated fluence maps. The reconstructed dose distribution, dosimetric endpoints, and DVHs were compared with that of the treatment plan.

RESULTS

Phantom study showed that HN CBCT can be directly used for dose reconstruction. For most treatment sessions, the CBCT-based dose reconstructions yielded DVHs of the targets close (within 3%) to that of the original treatment plans. However, dosimetric changes (within 10%) due to anatomic variations caused by setup inaccuracy, organ deformation, tumour shrinkage, or weight loss (or a combination of these) were observed for the critical organs.

CONCLUSIONS

The methodology we established affords an objective dosimetric basis for the clinical decision on whether a replanning is necessary during the course of treatment and provides a valuable platform for adaptive therapy in future.