Incorporating biological modeling into patient-specific plan verification.

PURPOSE: Dose-volume histogram (DVH) measurements have been integrated into commercially available quality assurance systems to provide a metric for evaluating accuracy of delivery in addition to gamma analysis. We hypothesize that tumor control probability and normal tissue complication probability calculations can provide additional insight beyond conventional dose delivery verification methods. METHODS: A commercial quality assurance system was used to generate DVHs of treatment plan using the planning CT images and patient-specific QA measurements on a phantom. Biological modeling was performed on the DVHs produced by both the treatment planning system and the quality assurance system. RESULTS: The complication-free tumor control probability, P+ , has been calculated for previously treated intensity modulated radiotherapy (IMRT) patients with diseases in the following sites: brain (-3.9% ± 5.8%), head-neck (+4.8% ± 8.5%), lung (+7.8% ± 1.3%), pelvis (+7.1% ± 12.1%), and prostate (+0.5% ± 3.6%). CONCLUSION: Dose measurements on a phantom can be used for pretreatment estimation of tumor control and normal tissue complication probabilities. Results in this study show how biological modeling can be used to provide additional insight about accuracy of delivery during pretreatment verification.

Evaluation of the Elekta Agility MLC performance using high-resolution log files.

PURPOSE: With the advent of volumetric modulated arc therapy (VMAT) and intensity-modulated radiation therapy (IMRT) treatment techniques, the requirement for more elaborate approaches in reviewing linac components' integrity has become even more stringent. A possible solution to this challenge is to employ the usage of log files generated during treatment. The log files generated by the new generation of Elekta linacs record events at a higher frequency (25 Hz) than their predecessors, which allows for retrospective analysis and identification of subtle changes and provides another means of quality assurance. The ability to track machine components based on log files for each treatment can allow for constant monitoring of fraction consistency in addition to machine reliability. Using Elekta Agility log files, a set of tests were developed to evaluate the reliability and robustness of the multileaf collimators (MLCs). METHODS: To evaluate Elekta log file utilization for linac MLC QA effectiveness, five MLC test patterns were constructed to review the effects of leaf velocity and acceleration on positional accuracy, including gravitational effects for the Elekta MLC system. Each test was run five times in a particular setting to obtain reproducibility data and statistical averages. This study was performed on two identical Versa HD machines, each delivering a full set of test plans with all possible variations. Plans were delivered using Elekta's iCOMcat software and recorded log files were extracted. Log files were reformatted for readability and automatically analyzed in Matlab® . RESULTS: The Elekta Agility MLC system was shown to be capable of obtaining speeds within the range of 5-35 mm/s. MLC step and shoot tests have demonstrated the MLC system's capability of having positional repeatability, averaging 0.03- and 0.08-mm offsets with and without gravitational effects, respectively. The IMRT-specific tests have shown that gravitational effects are negligible with all positional tests averaging 0.5-mm offsets. The largest speed root-mean-square error (RMSE) for the MLC system was found at the maximum speed of 35 mm/s with an average error of 0.8 mm. For slower speeds, the value was found to be much lower. CONCLUSION: Utilizing log files has demonstrated the feasibility for higher precision of MLC motions to be reviewed, based on the performance tests that were instituted. Log files provide insight on the effects of friction, acceleration, and gravity, with MU's delivered that previously could not be reviewed in such detail. Based on our results, log file-based QA has enhanced our ability to review performance, functionality, and perform QA on Elekta's MLC system.

Investigation of error detection capabilities of phantom, EPID and MLC log file based IMRT QA methods.

A patient specific quality assurance (QA) should detect errors that originate anywhere in the treatment planning process. However, the increasing complexity of treatment plans has increased the need for improvements in the accuracy of the patient specific pretreatment verification process. This has led to the utilization of higher resolution QA methods such as the electronic portal imaging device (EPID) as well as MLC log files and it is important to know the types of errors that can be detected with these methods. In this study, we will compare the ability of three QA methods (Delta4 ®, MU-EPID, Dynalog QA) to detect specific errors. Multileaf collimator (MLC) errors, gantry angle, and dose errors were introduced into five volumetric modulated arc therapy (VMAT) plans for a total of 30 plans containing errors. The original plans (without errors) were measured five times with each method to set a threshold for detectability using two standard deviations from the mean and receiver operating characteristic (ROC) derived limits. Gamma passing percentages as well as percentage error of planning target volume (PTV) were used for passing determination. When applying the standard 95% pass rate at 3%/3 mm gamma analysis errors were detected at a rate of 47, 70, and 27% for the Delta4 , MU-EPID and Dynalog QA respectively. When using thresholds set at 2 standard deviations from our base line measurements errors were detected at a rate of 60, 30, and 47% for the Delta4 , MU-EPID and Dynalog QA respectively. When using ROC derived thresholds errors were detected at a rate of 60, 27, and 47% for the Delta4 , MU-EPID and Dynalog QA respectively. When using dose to the PTV and the Dynalog method 11 of the 15 small MLC errors were detected while none were caught using gamma analysis. A combination of the EPID and Dynalog QA methods (scaling Dynalog doses using EPID images) matches the detection capabilities of the Delta4 by adding additional comparison metrics. These additional metrics are vital in relating the QA measurement to the dose received by the patient which is ultimately what is being confirmed.

Anatomy-based, patient-specific VMAT QA using EPID or MLC log files.

In this project, we investigated the use of an electronic portal imaging device (EPID), together with the treatment planning system (TPS) and MLC log files, to determine the delivered doses to the patient and evaluate the agreement between the treatment plan and the delivered dose distribution. The QA analysis results are presented for 15 VMAT patients using the EPID measurements, the ScandiDos Delta4 dosimeter, and the beam fluence calculated from the multileaf collimator (MLC) log file. EPID fluence images were acquired in continuous acquisition mode for each of the patients and they were processed through an in-house MATLAB program to create an opening density matrix (ODM), which was used as the input fluence for the dose calculation in the TPS (Pinnacle3). The EPID used in this study was the aSi1000 Varian on a Novalis TX linac equipped with high-definition MLC. The actual MLC positions and gantry angles were retrieved from the MLC log files and the data were used to calculate the delivered dose distributions in Pinnacle. The resulting dose distributions were then compared against the corresponding planned dose distributions using the 3D gamma index with 3 mm/3% passing criteria. The ScandiDos Delta4 phantom was also used to measure a 2D dose distribution for all the 15 patients and a 2D gamma was calculated for each patient using the Delta4 software. The average 3D gamma using the EPID images was 96.1% ± 2.2%. The average 3D gamma using the log files was 98.7% ± 0.5%. The average 2D gamma from the Delta4 was 98.1% ± 2.1%. Our results indicate that the use of the EPID, combined with MLC log files and a TPS, is a viable method for QA of VMAT plans.