Incorporation of realistic delivery limitations into dynamic MLC treatment delivery

The first investigation of spot-scanning proton arc (SPArc) delivery time and accuracy with different delivery tolerance window settings

Objective. To investigate the impact of various delivery tolerance window settings on the treatment delivery time and dosimetric accuracy of spot-scanning proton arc (SPArc) therapy.Approach. SPArc plans were generated for three representative disease sites (brain, lung, and liver cancer) with an angle sampling frequency of 2.5°. An in-house dynamic arc controller was used to simulate the arc treatment delivery with various tolerance windows (±0.25, ±0.5, ±1, and ±1.25°). The controller generates virtual logfiles during the arc delivery simulation, such as gantry speed, acceleration and deceleration, spot position, and delivery sequence, similar to machine logfiles. The virtual logfile was then imported to the treatment planning system to reconstruct the delivered dose distribution and compare it to the initial SPArc nominal plan. A three-dimensional gamma index was used to quantitatively assess delivery accuracy. Total treatment delivery time and relative lost time (dynamic arc delivery time-fix beam delivery time)/fix beam delivery time) were reported.Main Results. The 3D gamma passing rate (GPR) was greater than 99% for all cases when using 3%/3 mm and 2%/2 mm criteria and the GPR (1%/1 mm criteria) degraded as the tolerance window opens. The total delivery time for dynamic arc delivery increased with the decreasing delivery tolerance window length. The average delivery time and the relative lost time (%) were 630 ± 212 s (253% ± 68%), 322 ± 101 s (81% ± 31%), 225 ± 60 s (27% ± 16%), 196 ± 41 s (11% ± 6%), 187 ± 29 s (6% ± 1%) for tolerance windows ±0.25, ±0.5, ±1, and ±1.25° respectively.Significance. The study quantitatively analyzed the dynamic SPArc delivery time and accuracy with different delivery tolerance window settings, which offer a critical reference in the future SPArc plan optimization and delivery controller design.

Comparison of MLC positioning deviations using log files and establishment of specific assessment parameters for different accelerators with IMRT and VMAT

Background and purpose

The study evaluated the differences in leaf positioning deviations by the log files of three advanced accelerators with two delivery techniques, and established specific assessment parameters of leaf positioning deviations for different types of accelerators.

Methods

A total of 420 treatment plans with 5 consecutive treatment log files were collected from the Trilogy, TrueBeam and Halcyon accelerators. Millennium MLC was equipped on the Trilogy and TrueBeam accelerators. A jawless design and dual-layer MLC were adopted on the Halcyon accelerator. 70 IMRT and 70 VMAT plans were selected randomly on each accelerator. The treatment sites of all plans included head and neck, chest, breast, pelvis and other sites. The parsing tasks for 2100 log files were proceeded by SunCheck software from Sun Nuclear Corporation. The maximum leaf root mean square (RMS) errors, 95th percentile errors and percentages of different leaf positioning errors were statistically analyzed. The correlations between these evaluation parameters and accelerator performance parameters (maximum leaf speed, mean leaf speed, gantry and arc angle) were analyzed.

Results

The average maximum leaf RMS errors of the Trilogy in the IMRT and VMAT plans were 0.44 ± 0.09 mm and 0.79 ± 0.07 mm, respectively, which were higher than the TrueBeam's 0.03 ± 0.01 mm, 0.03 ± 0.01 mm and the Halcyon's 0.05 ± 0.01 mm, 0.07 ± 0.01 mm. Similar data results were shown in the 95th percentile error. The maximum leaf RMS errors were strongly correlated with the 95th percentile errors (Pearson index > 0.5). The leaf positioning deviations in VMAT were higher than those in IMRT for all accelerators. In TrueBeam and Halcyon, leaf position errors above 1 mm were not found in IMRT and VMAT plans. The main influencing factor of leaf positioning deviation was the leaf speed, which has no strong correlation with gantry and arc angles.

Conclusions

Compared with the quality assurance guidelines, the MLC positioning deviations tolerances of the three accelerators should be tightened. For both IMRT and VMAT techniques, the 95th percentile error and the maximum RMS error are suggested to be tightened to 1.5 and 1 mm respectively for the Trilogy accelerator. In TrueBeam and Halcyon accelerators, the 95th percentile error and maximum RMS error of 1 and 0.5 mm, respectively, are considered appropriate.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13014-022-02097-0.

A tool for patient-specific prediction of delivery discrepancies in machine parameters using trajectory log files

PURPOSE

Multileaf collimator (MLC) delivery discrepancy between planned and actual (delivered) positions have detrimental effect on the accuracy of dose distributions for both IMRT and VMAT. In this study, we evaluated the consistency of MLC delivery discrepancies over the course of treatment and over time to verify that a predictive machine learning model would be applicable throughout the course of treatment. Next, the MLC and gantry positions recorded in prior trajectory log files were analyzed to build a machine learning algorithm to predict MLC positional discrepancies during delivery for a new treatment plan. An open source tool was developed and released to predict the MLC positional discrepancies at treatment delivery for any given plan.

METHODS

Trajectory log files of 142 IMRT plans and 125 VMAT plans from 9 Varian TrueBeam linear accelerators were collected and analyzed. The consistency of delivery discrepancy over patient-specific quality assurance (QA) and patient treatment deliveries was evaluated. Data were binned by treatment site and machine type to determine their relationship with MLC and gantry angle discrepancies. Motion-related parameters including MLC velocity, MLC acceleration, control point, dose rate, and gravity vector, gantry velocity and gantry acceleration, where applicable, were analyzed to evaluate correlations with MLC and gantry discrepancies. Several regression models, such as simple/multiple linear regression, decision tree, and ensemble method (boosted tree and bagged tree model) were used to develop a machine learning algorithm to predict MLC discrepancy based on MLC motion parameters.

RESULTS

MLC discrepancies at patient-specific QA differed from those at patient treatment deliveries by a small (mean = 0.0021 ± 0.0036 mm, P = 0.0089 for IMRT; mean = 0.0010 ± 0.0016 mm, P = 0.0003 for VMAT) but statistically significant amount, likely due to setting the gantry angle to zero for QA in IMRT. MLC motion parameters, MLC velocity and gravity vector, showed significant correlation (P < 0.001) with MLC discrepancy, especially MLC velocity, which had an approximately linear relationship (slope = -0.0027, P < 0.001, R²  = 0.79). Incorporating MLC motion parameters, the final generalized model trained by data from all linear accelerators can predict MLC discrepancy to a high degree of accuracy with high correlation (R²  = 0.86) between predicted and actual MLC discrepancies. The same prediction results were found across different treatment sites and linear accelerators.

CONCLUSION

We have developed a machine learning model using trajectory log files to predict the MLC discrepancies during delivery. This model has been a released as a research tool in which a DICOM-RT with predicted MLC positions can be generated using the original DICOM-RT file as input. This tool can be used to simulate radiotherapy treatment delivery and may be useful for studies evaluating plan robustness and dosimetric uncertainties from treatment delivery.

Optical and x-ray technology synergies enabling diagnostic and therapeutic applications in medicine

X-ray and optical technologies are the two central pillars for human imaging and therapy. The strengths of x-rays are deep tissue penetration, effective cytotoxicity, and the ability to image with robust projection and computed-tomography methods. The major limitations of x-ray use are the lack of molecular specificity and the carcinogenic risk. In comparison, optical interactions with tissue are strongly scatter dominated, leading to limited tissue penetration, making imaging and therapy largely restricted to superficial or endoscopically directed tissues. However, optical photon energies are comparable with molecular energy levels, thereby providing the strength of intrinsic molecular specificity. Additionally, optical technologies are highly advanced and diversified, being ubiquitously used throughout medicine as the single largest technology sector. Both have dominant spatial localization value, achieved with optical surface scanning or x-ray internal visualization, where one often is used with the other. Therapeutic delivery can also be enhanced by their synergy, where radio-optical and optical-radio interactions can inform about dose or amplify the clinical therapeutic value. An emerging trend is the integration of nanoparticles to serve as molecular intermediates or energy transducers for imaging and therapy, requiring careful design for the interaction either by scintillation or Cherenkov light, and the nanoscale design is impacted by the choices of optical interaction mechanism. The enhancement of optical molecular sensing or sensitization of tissue using x-rays as the energy source is an important emerging field combining x-ray tissue penetration in radiation oncology with the molecular specificity and packaging of optical probes or molecular localization. The ways in which x-rays can enable optical procedures, or optics can enable x-ray procedures, provide a range of new opportunities in both diagnostic and therapeutic medicine. Taken together, these two technologies form the basis for the vast majority of diagnostics and therapeutics in use in clinical medicine.

[Effects of Low MU in Respiratory Gated IMRT on MLC Position Accuracy and Dose Distribution]

PURPOSE

The purpose of this research is to clarify the effects of low monitor unit (MU) on multileaf collimator (MLC) position accuracy and dose distribution in intensity modulated radiotherapy (IMRT) using respiratory gated.

METHOD

In the phantom experiment, irradiation without respiratory gated and respiratory gated with low MU (3, 5, and 7 MU) were performed, and positional accuracy and dose distribution of MLC were analyzed. MLC positional accuracy was calculated from the log-files and the MLC position error, gap size error, MLC leaf speed were calculated and compared with the planned value. Gamma analysis of the dose distribution obtained from the irradiated films and the dose distribution of the treatment plans were carried out.

RESULTS

Without respiratory gated and respiratory gated, the frequency of gap size error that did not exceed 0.2 mm were more than 93% under all conditions. MLC position error increased with increasing MLC leaf speed. The determination coefficient of respiratory gated irradiation was lower by about 20% compared with that without respiratory gated, and variation from the approximate straight line occurs. The output difference due to low MU irradiation during respiratory gated was within 1% of the planned value. Although, the pass rate of gamma analysis differed in tumor size, the dose distribution well conformity at 96% or more for both without respiratory gated and respiratory gated. However, in the comparison of the profile in the MLC movement direction, respiratory gated irradiation at 3 MU showed a difference of about 9% at the edge of the irradiated field and about 6% at the point where the dose rapidly changed.

CONCLUSION

It was shown that MLC position accuracy due to stop and go of MLC leaf can be secured even with low MU irradiation of about 3 MU. However, attention should be paid to the dose of risk organs adjacent to the tumor margin.

Commissioning and quality assurance for VMAT delivery systems: An efficient time-resolved system using real-time EPID imaging

PURPOSE

An ideal commissioning and quality assurance (QA) program for Volumetric Modulated Arc Therapy (VMAT) delivery systems should assess the performance of each individual dynamic component as a function of gantry angle. Procedures within such a program should also be time-efficient, independent of the delivery system and be sensitive to all types of errors. The purpose of this work is to develop a system for automated time-resolved commissioning and QA of VMAT control systems which meets these criteria.

METHODS

The procedures developed within this work rely solely on images obtained, using an electronic portal imaging device (EPID) without the presence of a phantom. During the delivery of specially designed VMAT test plans, EPID frames were acquired at 9.5 Hz, using a frame grabber. The set of test plans was developed to individually assess the performance of the dose delivery and multileaf collimator (MLC) control systems under varying levels of delivery complexities. An in-house software tool was developed to automatically extract features from the EPID images and evaluate the following characteristics as a function of gantry angle: dose delivery accuracy, dose rate constancy, beam profile constancy, gantry speed constancy, dynamic MLC positioning accuracy, MLC speed and acceleration constancy, and synchronization between gantry angle, MLC positioning and dose rate. Machine log files were also acquired during each delivery and subsequently compared to information extracted from EPID image frames.

RESULTS

The largest difference between measured and planned dose at any gantry angle was 0.8% which correlated with rapid changes in dose rate and gantry speed. For all other test plans, the dose delivered was within 0.25% of the planned dose for all gantry angles. Profile constancy was not found to vary with gantry angle for tests where gantry speed and dose rate were constant, however, for tests with varying dose rate and gantry speed, segments with lower dose rate and higher gantry speed exhibited less profile stability. MLC positional accuracy was not observed to be dependent on the degree of interdigitation. MLC speed was measured for each individual leaf and slower leaf speeds were shown to be compensated for by lower dose rates. The test procedures were found to be sensitive to 1 mm systematic MLC errors, 1 mm random MLC errors, 0.4 mm MLC gap errors and synchronization errors between the MLC, dose rate and gantry angle controls systems of 1°. In general, parameters measured by both EPID and log files agreed with the plan, however, a greater average departure from the plan was evidenced by the EPID measurements.

CONCLUSION

QA test plans and analysis methods have been developed to assess the performance of each dynamic component of VMAT deliveries individually and as a function of gantry angle. This methodology relies solely on time-resolved EPID imaging without the presence of a phantom and has been shown to be sensitive to a range of delivery errors. The procedures developed in this work are both comprehensive and time-efficient and can be used for streamlined commissioning and QA of VMAT delivery systems.

The impact of different leaf limited accelerations on leaf-positional error of a self-developed MLC prototype

Multi-leaf collimator (MLC) provides an available approach to achieve sophisticated radiotherapy techniques and plays an important role in advanced high quality radiation therapy. Delivery accuracy of radiation therapy mainly depends on the leaf sequence generated by treatment plans and dynamic performance of MLCs. Previous studies have explored the influence of various leaf limited velocities and dose rates on the leaf-positional error. However, there is a lack of research on the influence of different leaf limited accelerations on the delivery accuracy or leaf-positional error. Confining our study to the technique of sliding window, we investigate the effect of various leaf limited accelerations in treatment plans on leaf-positional error of a self-developed MLC prototype. In order to preliminarily verify the assumption, experiments of leaf trajectories sequenced with different limited accelerations under the same displacement were performed. Thereafter, according to the principle of the sliding window technique and the prescribed intensity profile, leaf trajectories for different leaf limited accelerations from 1.0 m/s2 to 5.0 m/s2 were calculated. Conducted on the self-developed MLC prototype, the experiment results have shown that the maximum leaf-positional error can be kept within 0.5 mm while the average and RMS of the leaf-positional error vary from leaf limited accelerations. Under the proposal leaf limited acceleration of 3 m/s2, the self-developed MLC achieved the optimal dynamic performance. In conclusion, for commercial MLCs, optimal constraint for treatment plans should be investigated and determined because incorporation of reasonable servo-mechanical constraints into leaf sequencing algorithm can enhance the conformity between planned and delivered fields.

Evaluation of MLC performance in VMAT and dynamic IMRT by log file analysis

PURPOSE

This multi-institution study assessed the positioning accuracy of multileaf collimators (MLC) by analyzing log files. It determined the main machine parameters that affect MLC positioning errors for pre-TrueBeam (Clinac) and TrueBeam linacs.

METHODS

Around 30,000 dIMRT and VMAT log files belonging to 6 linacs from 4 different centers were analyzed. An in-house software was used to calculate 95th percentile and RMS error values and their correlation with certain parameters such as maximum leaf speed, mean leaf speed and gantry angle. The effect of MLC communication delay on error statistics was assessed in Clinac linacs. To that end MLC positioning error statistics were calculated with and without the delay effect.

RESULTS

For dIMRT treatments in Clinac linacs the mean leaf RMS error was 0.306mm with and 0.030mm without the delay effect. Leaf RMS error was closely linked to maximum and mean leaf speeds, but without the delay effect that link was weaker. No trend was observed between bank RMS error and gantry angle. Without the delay effect larger bank RMS errors were obtained for gantry angles with leaf movements against gravity. For VMAT treatments in TrueBeam linacs the mean leaf RMS error was 0.038mm. A link was also observed between leaf RMS error and maximum and mean leaf speeds.

CONCLUSION

TrueBeam MLC positioning errors are substantially lower than those of Clinac linacs. In Clinac machines the analysis of dynalogs without the delay effect allows us to study the influence of factors that are masked by the delay effect.

[Consideration of Dose Error in Dynamic MLC IMRT Using MLC Speed Control with Dose Rate Change]

In dynamic multi-leaf collimator (MLC) intensity-modulated radiotherapy (IMRT), the accuracy of delivered dose is influenced by the positional accuracy of the moving MLC. In order to assess the accuracy of the delivered dose during dynamic MLC IMRT, the delivered dose error in dynamic MLC IMRT using the MLC speed control with dose rate change was investigated. Sweeping gap sequence irradiation was performed with constant MLC leaf speed at 0.6 to 5 cm/s or changed MLC speed (4 steps). The positional accuracy of the moving MLC was evaluated from the trajectory log file. Absorbed dose measurements with sweeping field (Field size: 10 cm×10 cm, MLC leaf speed: 0.6 to 2.7 cm/s, MLC leaf gap width: 0.2 to 2.0 cm) were performed. The delivered dose error at each gap width was evaluated according to MLC leaf speed change. MLC positional errors and changes in delivered dose according to MLC leaf speed were within 0.07 mm and 0.6%, respectively, for all measurements. Beam hold-off did not occur under any conditions. We confirmed that TrueBeam can regulate MLC leaf speed below the maximum limit (2.5 cm/s) by changing the dose rate in real-time during irradiation in dynamic MLC IMRT. MLC gap error during irradiation was estimated within 0.2 mm at the maximum dose rate from the results of absolute dose measurements using dynamic MLC irradiation. In conclusion, TrueBeam can use the maximum dose rate for the treatment planning of dynamic MLC IMRT, which has an advantage of shorter treatment time.

An EPID-based system for gantry-resolved MLC quality assurance for VMAT

Multileaf collimator (MLC) positions should be precisely and independently mea-sured as a function of gantry angle as part of a comprehensive quality assurance (QA) program for volumetric-modulated arc therapy (VMAT). It is also ideal that such a QA program has the ability to relate MLC positional accuracy to patient-specific dosimetry in order to determine the clinical significance of any detected MLC errors. In this work we propose a method to verify individual MLC trajectories during VMAT deliveries for use as a routine linear accelerator QA tool. We also extend this method to reconstruct the 3D patient dose in the treatment planning sys-tem based on the measured MLC trajectories and the original DICOM plan file. The method relies on extracting MLC positions from EPID images acquired at 8.41fps during clinical VMAT deliveries. A gantry angle is automatically tagged to each image in order to obtain the MLC trajectories as a function of gantry angle. This analysis was performed for six clinical VMAT plans acquired at monthly intervals for three months. The measured trajectories for each delivery were compared to the MLC positions from the DICOM plan file. The maximum mean error detected was 0.07 mm and a maximum root-mean-square error was 0.8 mm for any leaf of any delivery. The sensitivity of this system was characterized by introducing random and systematic MLC errors into the test plans. It was demonstrated that the system is capable of detecting random and systematic errors on the range of 1-2mm and single leaf calibration errors of 0.5 mm. The methodology developed in the work has potential to be used for efficient routine linear accelerator MLC QA and pretreatment patient-specific QA and has the ability to relate measured MLC positional errors to 3D dosimetric errors within a patient volume.

A multi-institution evaluation of MLC log files and performance in IMRT delivery

Background

The multileaf collimator (MLC) is a critical component to accurate intensity-modulated radiotherapy (IMRT) delivery. This study examined MLC positional accuracy via MLC logs from Varian machines from six institutions and three delivery techniques to evaluate typical positional accuracy and treatment and mechanical parameters that affect accuracy. Typical accuracy achieved was compared against TG-142 recommendations for MLC performance; more appropriate recommendations are suggested.

Methods

Over 85,000 Varian MLC treatment logs were collected from six institutions and analyzed with FractionCHECK. Data were binned according to institution and treatment type to determine overall root mean square (RMS) and 95^th percentile error values, and then to look for correlations between those errors and with mechanical and treatment parameters including mean and maximum leaf speed, gantry angle, beam-on time, mean leaf error, and number of segments.

Results

Results of treatment logs found that leaf RMS error and 95^th percentile leaf error were consistent between institutions, but varied by treatment type. The step and shoot technique had very small errors: the mean RMS leaf error was 0.008 mm. For dynamic treatments the mean RMS leaf error was 0.32 mm, while volumetric-modulated arc treatment (VMAT) showed an RMS leaf error of 0.46 mm. Most MLC leaf errors were found to be well below TG-142 recommended tolerances. For the dynamic and VMAT techniques, the mean and maximum leaf speeds were significantly linked to the leaf RMS error. Additionally, for dynamic delivery, the mean leaf error was correlated with RMS error, whereas for VMAT the average gantry speed was correlated. For all treatments, the RMS error and the 95^th percentile leaf error were correlated.

Conclusions

Restricting the maximum leaf speed can help improve MLC performance for dynamic and VMAT deliveries. Furthermore, the tolerances of leaf RMS and error counts for all treatment types should be tightened from the TG-142 values to make them more appropriate for clinical performance. Values of 1 mm for the 95^th percentile of leaf RMS error and 1.5 mm for the 95^th percentile leaf error are suggested as action levels for all treatment types.

Use of plan quality degradation to evaluate tradeoffs in delivery efficiency and clinical plan metrics arising from IMRT optimizer and sequencer compromises

PURPOSE

Plan degradation resulting from compromises made to enhance delivery efficiency is an important consideration for intensity modulated radiation therapy (IMRT) treatment plans. IMRT optimization and/or multileaf collimator (MLC) sequencing schemes can be modified to generate more efficient treatment delivery, but the effect those modifications have on plan quality is often difficult to quantify. In this work, the authors present a method for quantitative assessment of overall plan quality degradation due to tradeoffs between delivery efficiency and treatment plan quality, illustrated using comparisons between plans developed allowing different numbers of intensity levels in IMRT optimization and/or MLC sequencing for static segmental MLC IMRT plans.

METHODS

A plan quality degradation method to evaluate delivery efficiency and plan quality tradeoffs was developed and used to assess planning for 14 prostate and 12 head and neck patients treated with static IMRT. Plan quality was evaluated using a physician's predetermined "quality degradation" factors for relevant clinical plan metrics associated with the plan optimization strategy. Delivery efficiency and plan quality were assessed for a range of optimization and sequencing limitations. The "optimal" (baseline) plan for each case was derived using a clinical cost function with an unlimited number of intensity levels. These plans were sequenced with a clinical MLC leaf sequencer which uses >100 segments, assuring delivered intensities to be within 1% of the optimized intensity pattern. Each patient's optimal plan was also sequenced limiting the number of intensity levels (20, 10, and 5), and then separately optimized with these same numbers of intensity levels. Delivery time was measured for all plans, and direct evaluation of the tradeoffs between delivery time and plan degradation was performed.

RESULTS

When considering tradeoffs, the optimal number of intensity levels depends on the treatment site and on the stage in the process at which the levels are limited. The cost of improved delivery efficiency, in terms of plan quality degradation, increased as the number of intensity levels in the sequencer or optimizer decreased. The degradation was more substantial for the head and neck cases relative to the prostate cases, particularly when fewer than 20 intensity levels were used. Plan quality degradation was less severe when the number of intensity levels was limited in the optimizer rather than the sequencer.

CONCLUSIONS

Analysis of plan quality degradation allows for a quantitative assessment of the compromises in clinical plan quality as delivery efficiency is improved, in order to determine the optimal delivery settings. The technique is based on physician-determined quality degradation factors and can be extended to other clinical situations where investigation of various tradeoffs is warranted.

Performance evaluation of respiratory motion-synchronized dynamic IMRT delivery

The purpose of this study was to evaluate the capabilities of DMLC to deliver the respiratory motion‐synchronized dynamic IMRT (MS‐IMRT) treatments under various dose rates. In order to create MS‐IMRT plans, the DMLC leaf motions in dynamic IMRT plans of eight lung patients were synchronized with the respiratory motion of breathing period 4 sec and amplitude 2 cm (peak to peak) using an in‐house developed leaf position modification program. The MS‐IMRT plans were generated for the dose rates of 100 MU/min, 400 MU/min, and 600 MU/min. All the MS‐IMRT plans were delivered in a medical linear accelerator, and the fluences were measured using a 2D ion chamber array, placed over a moving platform. The accuracy of MS‐IMRT deliveries was evaluated with respect to static deliveries (no compensation for target motion) using gamma test. In addition, the fluences of gated delivery of 30% duty cycle and non‐MS‐IMRT deliveries were also measured and compared with static deliveries. The MS‐IMRT was better in terms of dosimetric accuracy, compared to gated and non‐MS‐IMRT deliveries. The dosimetric accuracy was observed to be significantly better for 100 MU/min MS‐IMRT. However, the use of high‐dose rate in a MS‐IMRT delivery introduced dose‐rate modulation/beam hold‐offs that affected the synchronization between the DMLC leaf motion and target motion. This resulted in more dose deviations in MS‐IMRT deliveries at the dose rate of 600 MU/min.

PACS numbers: 87.53.kn, 87.56.N‐

Quality, technology and outcomes: evolution and evaluation of new treatments and/or new technology

The pace of technological innovation and adoption continues to increase each year, and the field of Radiation Oncology struggles to react appropriately to the changes and potential improvements in treatment which hopefully will result from this innovation. The standard methods used in the past to test new technology and treatment techniques are often no longer appropriate for this fast-changing environment. This paper uses examples from radiotherapy technological developments over the last decades to illustrate issues which need to be solved in order to study and evaluate potential advances, and then describes several improved ways to study new techniques and technology. Design of appropriate studies can help us improve patient care while at the same time documenting which new clinical strategies, enabled by new technology, lead to improved patient outcomes.

The effect of gantry and collimator angles on leaf limited velocity and position in dynamic multileaf collimator intensity-modulated radiation therapy

The purpose of the study is to evaluate the limiting velocity (LV) of a multileaf collimator and the leaf position in various collimator and gantry angles. Both leading leaves and trailing leaves began to move with a constant acceleration from 0 to 4 cm s(-1). When the beam hold occurred, the leaf velocity was defined as the leaf LV. Dynamic irradiation was performed at eight gantry angles of every 45 degrees with three different collimator angles. The analysis of the LV and the leaf position was performed with a log file from a leaf motion controller. The mean LVs for Varian Clinac 21EX (21EX) ranged from 2.51 to 3.10 cm s(-1). The mean LVs for Clinac 600C ranged from 2.91 to 3.12 cm s(-1). When only central 5 mm leaves of 21EX moved, LVs were significantly higher than those when all 60 pairs of leaf moved, while the leaf position inconsistencies of the two accelerators were within 1 mm at the leaf velocities from 0.5 to 2.0 cm s(-1). It was recognized that the LV was affected by gravity. This measurement method can be utilized as routine quality assurance for a dynamic multileaf collimator (DMLC) is and easily reproducible.

Overestimation of low-dose radiation in intensity-modulated radiotherapy with sliding-window technique

PURPOSE

To analyze different control-system limitations on the measured dose distributions in low-dose regions of simplified intensity fields with an electronic portal imaging device to ascertain the optimal settings for the control-system limitations in the planning system.

MATERIAL AND METHODS

The authors created one field with an "optimal fluence" of intensity 1.0 (full dose) and one field with intensity 0.0 (no dose) in the central part of the field. The influence of different dose rates (DRs) and maximum leaf speeds (LS) on the calculated and measured dose and dose profiles were analyzed.

RESULTS

Good agreement between calculated and measured dose in the case of a field of intensity 1.0 was found. For the field with intensity 0.0, the measured dose was 20-60% lower than the dose calculated by the "actual fluence". The results were found dependent on the DR and LS.

CONCLUSION

The overestimation in regions of optimal intensity 0.0 by the planning system cannot be resolved by the user. Taking the measured dose in the region of desired intensity 1.0 and other technical limitations (like beam hold interrupts or spikes in the cross and longitudinal profiles) into consideration, the application of an LS of 2.5 cm/s and a DR of 500 MU/min is recommended in order to minimize radiation dose applied to organs at risk, which are located in regions of low intensity, like, for example, the spinal cord.

Impact of different leaf velocities and dose rates on the number of monitor units and the dose-volume-histograms using intensity modulated radiotherapy with sliding-window technique

Background

Intensity modulated radiotherapy (IMRT) using sliding window technique utilises a leaf sequencing algorithm, which takes some control system limitations like dose rates (DR) and velocity of the leafs (LV) into account. The effect of altering these limitations on the number of monitor units and radiation dose to the organs at risk (OAR) were analysed.

Methods

IMRT plans for different LVs from 1.0 cm/sec to 10.0 cm/sec and different DRs from 100 MU/min to 600 MU/min for two patients with prostate cancer and two patients with squamous cell cancer of the scalp (SCCscalp) were calculated using the same "optimal fluence map". For each field the number of monitor units, the dose volume histograms and the differences in the "actual fluence maps" of the fields were analysed.

Results

With increase of the DR and decrease of the LV the number of monitor units increased and consequentially the radiation dose given to the OAR. In particular the serial OARs of patients with SCCscalp, which are located outside the end position of the leafs and inside the open field, received an additional dose of a higher DR and lower LV is used.

Conclusion

For best protection of organs at risk, a low DR and high LV should be applied. But the consequence of a low DR is both a long treatment time and also that a LV of higher than 3.0 cm/sec is mechanically not applicable. Our recommendation for an optimisation of the discussed parameters is a leaf velocity of 2.5 cm/sec and a dose rate of 300–400 MU/min (prostate cancer) and 100–200 MU/min (SCCscalp) for best protection of organs at risk, short treatment time and number of monitor units.

Performance of a direct-detection active matrix flat panel dosimeter (AMFPD) for IMRT measurements

The dosimetric performance of a direct-detection active matrix flat panel dosimeter (AMFPD) is reported for intensity modulated radiation therapy (IMRT) measurements. The AMFPD consists of a-Si : H photodiodes and thin-film transistors deposited on a glass substrate with no overlying scintillator screen or metal plate. The device is operated at 0.8 frames per second in a continuous acquisition or fluoroscopic mode. The effect of the applied bias voltage across the photodiodes on the response of the AMFPD was evaluated because this parameter affects dark signal, lag contributions, and pixel sensitivity. In addition, the AMPFD response was evaluated as a function of dose, dose rate, and energy, for static fields at 10 cm depth. In continuous acquisition mode, the AMFPD maintained a linear dose response (r2 > 0.99999) up to at least 1040 cGy. In order to obtain reliable integrated dose results for IMRT fields, the effects of lag on the radiation signal were minimized by operating the system at the highest frame rate and at an appropriate reverse bias voltage. Segmental MLC and dynamic MLC IMRT fields were measured with the AMFPD, and the results were compared to film, using standard methods for reliable film dosimetry. Both AMFPD and film measurements were independently converted to dose in cGy. Gamma and chi values were calculated as indices of agreement. The results from the AMFPD were in excellent agreement with those from film. When 2% of D(max) and 2 mm of distance to agreement were used as the criteria, 98% of the region of interest (defined as the region where dose is greater than 5% of D(max)) satisfied [chi] < or = 1 on average across the cases that were tested.

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

Due to the complicated technical nature of gated radiation therapy, electronic and mechanical limitations may affect the precision of delivery. The purpose of this study is to investigate the geometric and dosimetric accuracies of gated step-and-shoot intensity modulated radiation treatments (SS-IMRT). Unique segmental MLC plans are designed, which allow quantitative testing of the gating process. Both ungated and gated deliveries are investigated for different dose sizes, dose rates, and gating window times using a commercial treatment system (Varian Trilogy) together with a respiratory gating system [Varian Real-Time Position Management system]. Radiographic film measurements are used to study the geometric accuracy, where it is found that with both ungated and gated SS-IMRT deliveries the MLC leaf divergence away from planned is less than or equal to the MLC specified leaf tolerance value for all leafs (leaf tolerance being settable from 0.5-5 mm). Nevertheless, due to the MLC controller design, failure to define a specific leaf tolerance value suitable to the SS-IMRT plan can lead to undesired geometric effects, such as leaf motion of up to the maximum 5 mm leaf tolerance value occurring after the beam is turned on. In this case, gating may be advantageous over the ungated case, as it allows more time for the MLC to reach the intended leaf configuration. The dosimetric precision of gated SS-IMRT is investigated using ionization chamber methods. Compared with the ungated case, it is found that gating generally leads to increased dosimetric errors due to the interruption of the "overshoot phenomena." With gating the average timing deviation for intermediate segments is found to be 27 ms, compared to 18 ms for the ungated case. For a plan delivered at 600 MU/min this would correspond to an average segment dose error of approximately 0.27 MU and approximately 0.18 MU for gated and ungated deliveries, respectively. The maximum dosimetric errors for individual intermediate segments are found to deviate by up to approximately 0.64 MU from their planned value when delivered at 600 MU/min using gating, this compares to only approximately 0.32 MU for the ungated case.

Computational Challenges for Image-Guided Radiation Therapy: Framework and Current Research

It is arguable that the imaging and delivery hardware necessary for delivering real-time adaptive image-guided radiotherapy is available on high-end linear accelerators. Robust and computationally efficient software is the limiting factor in achieving highly accurate and precise radiotherapy to the constantly changing anatomy of a cancer patient. The limitations are not caused by the availability of algorithms but rather issues of reliability, integration, and calculation time. However, each of the software components is an active area of research and development at academic and commercial centers. This article describes the software solutions in 4 broad areas: deformable image registration, adaptive replanning, real-time image guidance, and dose calculation and accumulation. Given the pace of technological advancement, the integration of these software solutions to develop real-time adaptive image-guided radiotherapy and the associated challenges they bring will be implemented to varying degrees by all major manufacturers over the coming years.

Experimental verification of a Monte Carlo-based MLC simulation model for IMRT dose calculation

Inter- and intra-leaf transmission and head scatter can play significant roles in intensity modulated radiation therapy (IMRT)-based treatment deliveries. In order to accurately calculate the dose in the IMRT planning process, it is therefore important that the detailed geometry of the multi-leaf collimator (MLC), in addition to other components in the accelerator treatment head, be accurately modeled. In this paper, we have used the Monte Carlo method (MC) to develop a comprehensive model of the Varian 120 leaf MLC and have compared it against measurements in homogeneous phantom geometries under different IMRT delivery circumstances. We have developed a geometry module within the DPM MC code to simulate the detailed MLC design and the collimating jaws. Tests consisting of leakage, leaf positioning and static MLC shapes were performed to verify the accuracy of transport within the MLC model. The calculations show agreement within 2% in the high dose region for both film and ion-chamber measurements for these static shapes. Clinical IMRT treatment plans for the breast [both segmental MLC (SMLC) and dynamic MLC (DMLC)], prostate (SMLC) and head and neck split fields (SMLC) were also calculated and compared with film measurements. Such a range of cases were chosen to investigate the accuracy of the model as a function of modulation in the beamlet pattern, beamlet width, and field size. The overall agreement is within 2% /2 mm of the film data for all IMRT beams except the head and neck split field, which showed differences up to 5% in the high dose regions. Various sources of uncertainties in these comparisons are discussed.

Synchronized dynamic dose reconstruction

Variations in target volume position between and during treatment fractions can lead to measurable differences in the dose distribution delivered to each patient. Current methods to estimate the ongoing cumulative delivered dose distribution make idealized assumptions about individual patient motion based on average motions observed in a population of patients. In the delivery of intensity modulated radiation therapy (IMRT) with a multi-leaf collimator (MLC), errors are introduced in both the implementation and delivery processes. In addition, target motion and MLC motion can lead to dosimetric errors from interplay effects. All of these effects may be of clinical importance. Here we present a method to compute delivered dose distributions for each treatment beam and fraction, which explicitly incorporates synchronized real-time patient motion data and real-time fluence and machine configuration data. This synchronized dynamic dose reconstruction method properly accounts for the two primary classes of errors that arise from delivering IMRT with an MLC: (a) Interplay errors between target volume motion and MLC motion, and (b) Implementation errors, such as dropped segments, dose over/under shoot, faulty leaf motors, tongue-and-groove effect, rounded leaf ends, and communications delays. These reconstructed dose fractions can then be combined to produce high-quality determinations of the dose distribution actually received to date, from which individualized adaptive treatment strategies can be determined.

An experimental investigation into the radiation field offset of a dynamic multileaf collimator

In this study we investigate the characteristics of a rounded leaf end multileaf collimator (MLC) that is used for delivering intensity-modulated radiotherapy (IMRT) with a Varian linear accelerator. The rounded leaf end MLC design results in an offset between the radiation field edge (the physical leaf position) and the light field (the geometric leaf position). We call this the radiation field offset (RFO). The leaf position is calibrated to the leaf tip at the mid-leaf plane. There is an additional offset between the geometric leaf position and the projected leaf tip position that varies as a function of distance from the collimator central axis due to the MLC geometry. We call this the leaf position offset (LPO). There is a lack of consistency in the interpretation and implementation of the RFO and the LPO in the literature. We investigated the RFO and the LPO on Varian's 600 C/D and 21 EX linear accelerators. We used a combination of film and ion chamber measurements of static, segmental MLC (SMLC) and dynamic MLC (DMLC) fields to quantify the leaf offsets across the range of leaf positions. We were able to improve the dosimetry at large off-axis positions with minor adjustments to the vendor's LPO file. The RFO was determined to within 0.1 mm accuracy at the collimator central axis. The measured RFO value depends on whether the method is based on the radiation field edge position or on an integral dose measurement. The integral dose method results in an RFO that is approximately 0.2 mm greater than the radiation field edge method. The difference is due to the MLC penumbra shape. We propose a methodology for measuring and implementing MLC leaf offsets that is suitable for both SMLC and DMLC IMRT. In addition, we propose some definitions that more clearly describe the MLC leaf position for accurate IMRT dosimetry.

Measurement for the MLC leaf velocity profile by considering the leaf leakage using a radiographic film

A method to measure the velocity profile of a multi-leaf collimator (MLC) leaf along its travel range using a radiographic film is reported by considering the intra-leaf leakage. A specific dynamic MLC field with leaves travelling from the field edge to the isocentre line was designed. The field was used to expose a radiographic film, which was then scanned, and the dose profile along the horizontal leaf axis was measured. The velocity at a sampling point on the film can be calculated by considering the horizontal distance between the sampling point and the isocentre line, dose at the sampling point, dose rate of the linear accelerator, the total leaf travel time from the field edge to isocentre line and the pre-measured dose rate of leaf leakage. With the leaf velocities and velocity profiles for all MLC leaves measured routinely, a comprehensive and simple QA for the MLC can be set up to test the consistency of the leaf velocity performance which is essential to the IMRT delivery using a sliding window technique.

Monte Carlo based IMRT dose verification using MLC log files and R/V outputs

Conventional IMRT dose verification using film and ion chamber measurements is useful but limited with respect to the actual dose distribution received by the patient. The Monte Carlo simulation has been introduced as an independent dose verification tool for IMRT using the patient CT data and MLC leaf sequence files, which validates the dose calculation accuracy but not the plan delivery accuracy. In this work, we propose a Monte Carlo based IMRT dose verification method that reconstructs the patient dose distribution using the patient CT, actual beam data based on the information from the record and verify system (R/V), and the MLC log files obtained during dose delivery that record the MLC leaf positions and MUs delivered. Comparing the Monte Carlo dose calculation with the original IMRT plan using these data simultaneously validates the accuracy of both the IMRT dose calculation and beam delivery. Such log file based Monte Carlo simulations are expected to be employed as a useful and efficient IMRT QA modality to validate the dose delivered to the patient. We have run Monte Carlo simulations for eight IMRT prostate plans using this method and the results for the target dose were consistent with the original CORVUS treatment plans to within 3.0% and 2.0% with and without heterogeneity corrections in the dose calculation. However, significant dose deviations in nearby critical structures have been observed. The results showed that up to 9.0% of the bladder dose and up to 38.0% of the rectum dose, to which leaf position errors were found to contribute <2%, were underestimated by the CORVUS treatment planning system. The concept of average leaf position error has been defined to analyze MLC leaf position errors for an IMRT plan. A linear correlation between the target dose error and the average position error has been found based on log file based Monte Carlo simulations, showing that an average position error of 0.2 mm can result in a target dose error of about 1.0%.

SWIMRT: a graphical user interface using sliding window algorithm to construct a fluence map machine file

A custom‐made computer program, SWIMRT, to construct “multileaf collimator (MLC) machine” file for intensity‐modulated radiotherapy (IMRT) fluence maps was developed using MATLAB® and the sliding window algorithm. The user can either import a fluence map with a graphical file format created by an external treatment‐planning system such as Pinnacle3 or create his or her own fluence map using the matrix editor in the program. Through comprehensive calibrations of the dose and the dimension of the imported fluence field, the user can use associated image‐processing tools such as field resizing and edge trimming to modify the imported map. When the processed fluence map is suitable, a “MLC machine” file is generated for our Varian 21 EX linear accelerator with a 120‐leaf Millennium MLC. This machine file is transferred to the MLC console of the LINAC to control the continuous motions of the leaves during beam irradiation. An IMRT field is then irradiated with the 2D intensity profiles, and the irradiated profiles are compared to the imported or modified fluence map. This program was verified and tested using film dosimetry to address the following uncertainties: (1) the mechanical limitation due to the leaf width and maximum traveling speed, and (2) the dosimetric limitation due to the leaf leakage/transmission and penumbra effect. Because the fluence map can be edited, resized, and processed according to the requirement of a study, SWIMRT is essential in studying and investigating the IMRT technique using the sliding window algorithm. Using this program, future work on the algorithm may include redistributing the time space between segmental fields to enhance the fluence resolution, and readjusting the timing of each leaf during delivery to avoid small fields. Possible clinical utilities and examples for SWIMRT are given in this paper.

PACS numbers: 87.53.Kn, 87.53.St, 87.53.Uv

Verification of a rounded leaf-end MLC model used in a radiotherapy treatment planning system

A new multileaf collimator (MLC) model has been incorporated into version 7.4 of the Pinnacle radiotherapy treatment planning system (Philips Radiation Oncology Systems, Milpitas, CA). The MLC model allows for rounded MLC leaf-ends and provides separate parameters for inter-leaf transmission, intra-leaf transmission and the tongue width of the MLC leaf. In this report we detail the method followed to commission the MLC model for a Varian 120-leaf Millennium MLC (Varian Medical Systems, Palo Alto, CA, USA) for both 6 and 10 MV photons, and test the validity of the model for an IMRT field. Dose profiles in water were measured for a range of square MLC field sizes and compared to the Pinnacle computed dose profiles; in addition, the dose distribution for a series of adjacent MLC fields was measured to observe the model's behaviour along match-lines. Based on these results intra-leaf transmissions of 1.5% for 6 MV and 1.8% for 10 MV, leaf-tip radius of 12.0 cm, an inter-leaf transmission of 0.5%, and a tongue width of 0.1 cm were chosen. Using these values to compute the planar dose distribution for a 6 MV IMRT field, the new version of Pinnacle displayed improved dosimetric agreement with the dose-to-water EPID image and ion chamber measurements when compared to the old version of Pinnacle, particularly along the MLC tongue edge and across match-lines. Discrepancies of up to 5% were observed between calculated and measured doses along match-lines for both 6 MV and 10 MV photons; however, the new MLC model did predict the presence of match-lines and was a significant improvement on the previous model.

A dose gradient analysis tool for IMRT QA

The use of intensity-modulated radiation therapy (IMRT) has led to an increase in the number of complex fields that require measurement and comparison to calculated dose distributions in 2D. Current dose-evaluation techniques, including isodose line comparisons, displays of the dose difference between calculated and measured distributions, and distance-to-agreement (DTA) comparisons, are useful for display of differences between two different dose distributions but are often of limited value for the assessment of the discrepancies in terms of significance and/or cause. In this paper, we present a new gradient compensation method for the evaluation of local dosimetric differences as a function of the dose gradient at each point in the dose distribution. To apply the method, the user specifies a distance parameter (typically 1 mm), which is the geometric tolerance the user is prepared to accept for the dose comparison. The expected geometric uncertainties in the comparison process can include finite calculation and measurement grids, small misalignments of measured and calculated results, and volume-averaging effects in the measurement detector. Since these uncertainties can obscure the interpretation of any of the analysis tools described above, removing dose differences related to the tolerable geometric uncertainty helps the gradient compensation method highlight algorithm and delivery-related differences. The remaining dose differences not explained by the geometric tolerance can then be evaluated graphically (dose difference display) or analytically (dose difference dose-volume histograms)over the entire comparison region.

Maximum MLC opening effect in dynamic delivery of IMRT: leaf-positional analysis

The analysis of dynamic multileaf collimator (MLC) positions for the delivered intensity-modulated radiotherapy (IMRT) plans is crucial in that it may capture dose delivery problems otherwise difficult to observe and quantify in the conventional dosimetric measurements with film or with an ionization chamber. In some IMRT systems, delivery of IMRT fields starts with a maximum MLC opening (roughly the shape of the target in the beam's-eye view) and then proceeds to the subsequent dynamic MLC subfields. No irradiation is required in going from the initial segment (maximum opening) to the next one, and theoretically, no dose should be delivered in that initial moment. However, due to a finite sampling time of the MLC controller, the finite speed of the MLC, and a finite leaf tolerance, there may be some dose delivered between the first and the second segment. The amount of the excess dose is higher for larger dose rates and for a smaller number of the total monitor units per IMRT field. The magnitude of the dose errors could be in the order of a few percent. Effects similar to the maximum MLC opening may occur in other situations as well, for instance, when leaves are forced to move over large distances in a short time. Confounding this are dose errors due to the uncertainty in the MLC transmission. The analysis of the actual leaf positions recorded in the dynamic MLC log file is helpful in differentiating between the two types of errors and in determining the optimal dynamic MLC delivery parameters.

Improving IMRT delivery efficiency using intensity limits during inverse planning

Inverse planned intensity modulated radiotherapy (IMRT) fields can be highly modulated due to the large number of degrees of freedom involved in the inverse planning process. Additional modulation typically results in a more optimal plan, although the clinical rewards may be small or offset by additional delivery complexity and/or increased dose from transmission and leakage. Increasing modulation decreases delivery efficiency, and may lead to plans that are more sensitive to geometrical uncertainties. The purpose of this work is to assess the use of maximum intensity limits in inverse IMRT planning as a simple way to increase delivery efficiency without significantly affecting plan quality. Nine clinical cases (three each for brain, prostate, and head/neck) were used to evaluate advantages and disadvantages of limiting maximum intensity to increase delivery efficiency. IMRT plans were generated using in-house protocol-based constraints and objectives for the brain and head/neck, and RTOG 9406 dose volume objectives in the prostate. Each case was optimized at a series of maximum intensity ratios (the product of the maximum intensity and the number of beams divided by the prescribed dose to the target volume), and evaluated in terms of clinical metrics, dose-volume histograms, monitor units (MU) required per fraction (SMLC and DMLC delivery), and intensity map variation (a measure of the beam modulation). In each site tested, it was possible to reduce total monitor units by constraining the maximum allowed intensity without compromising the clinical acceptability of the plan. Monitor unit reductions up to 38% were observed for SMLC delivery, while reductions up to 29% were achieved for DMLC delivery. In general, complicated geometries saw a smaller reduction in monitor units for both delivery types, although DMLC delivery required significantly more monitor units in all cases. Constraining the maximum intensity in an inverse IMRT plan is a simple way to improve delivery efficiency without compromising plan objectives.

Determination of maximum leaf velocity and acceleration of a dynamic multileaf collimator: Implications for 4D radiotherapy

The dynamic multileaf collimator (MLC) can be used for four-dimensional (4D), or tumor tracking radiotherapy. However, the leaf velocity and acceleration limitations become a crucial factor as the MLC leaves need to respond in near real time to the incoming respiration signal. The aims of this paper are to measure maximum leaf velocity, acceleration, and deceleration to obtain the mechanical response times for the MLC, and determine whether the MLC is suitable for 4D radiotherapy. MLC leaf sequence files, requiring the leaves to reach maximum acceleration and velocity during motion, were written. The leaf positions were recorded every 50 ms, from which the maximum leaf velocity, acceleration, and deceleration were derived. The dependence on the velocity and acceleration of the following variables were studied: leaf banks, inner and outer leaves, MLC-MLC variations, gravity, friction, and the stability of measurements over time. Measurement results show that the two leaf banks of a MLC behave similarly, while the inner and outer leaves have significantly different maximum leaf velocities. The MLC-MLC variations and the dependence of gravity on maximum leaf velocity are statistically significant. The average maximum leaf velocity at the isocenter plane of the MLC ranged from 3.3 to 3.9 cm/s. The acceleration and deceleration at the isocenter plane of the MLC ranged from 50 to 69 cm/s2 and 46 to 52 cm/s2, respectively. Interleaf friction had a negligible effect on the results, and the MLC parameters remained stable with time. Equations of motion were derived to determine the ability of the MLC response to fluoroscopymeasured diaphragm motion. Given the present MLC mechanical characteristics, 4D radiotherapy is feasible for up to 97% of respiratory motion. For the largest respiratory motion velocities observed, beam delivery should be temporarily stopped (beam hold).

An extensive log-file analysis of step-and-shoot intensity modulated radiation therapy segment delivery errors

We present a study to evaluate the monitor unit (MU), dosimetric, and leaf-motion errors found in the delivery of 91 step-and-shoot IMRT treatment plans performed at three nominal dose rates using a dual modality high energy Linac (Varian 2100 C/D, Varian Medical Systems Inc., Palo Alto, CA) equipped with a 120-leaf multileaf collimator (MLC). The analysis was performed by studying log files generated by the MLC controller system. Recent studies by our group have validated that the automatically generated MLC log files accurately record the actual system delivery. A total of 635 beams were delivered at three nominal dose rates: 100, 300, and 600 MU/min. The log files were manually retrieved and analysis software was developed to extract the recorded MU delivery and leaf positions for each segment. Our analysis revealed that the magnitude of segment MU errors were independent of the planned segment MUs. Segment MU errors were found to increase with dose rate having maximum errors per segment of +/-1.8 MU at 600 MU/min, +/-0.8 MU at 300 MU/min, and +/-0.5 MU at 100 MU/min. The total absolute MU error in each plan was observed to increase with the number of plan segments, with the trend increasing more rapidly for higher dose rates. Three dimensional dose distributions were recomputed based on the observed segment MU errors for three plans with large cumulative absolute MU errors. Comparison with the original treatment plans indicated no clinically significant consequences due to these errors. In addition, approximately 80% of the total segment deliveries reported at least one collimator leaf moving at least 1 mm (projected at isocenter) during segment delivery. Such errors occur near the end of segment delivery and have been previously observed by our group using a fast video-based electronic portal imaging device. At 600 MU/min, between 5% and 23% of the plan MUs were delivered during leaf motion that had exceeded a 1 mm position tolerance. These leaf motion errors were not included in the treatment plan recalculations performed in this study.

Dysphagia and aspiration after chemoradiotherapy for head-and-neck cancer: which anatomic structures are affected and can they be spared by IMRT?

PURPOSE

To identify the anatomic structures whose damage or malfunction cause late dysphagia and aspiration after intensive chemotherapy and radiotherapy (RT) for head-and-neck cancer, and to explore whether they can be spared by intensity-modulated RT (IMRT) without compromising target RT.

METHODS AND MATERIALS

A total of 26 patients receiving RT concurrent with gemcitabine, a regimen associated with a high rate of late dysphagia and aspiration, underwent prospective evaluation of swallowing with videofluoroscopy (VF), direct endoscopy, and CT. To assess whether the VF abnormalities were regimen specific, they were compared with the VF findings of 6 patients presenting with dysphagia after RT concurrent with high-dose intra-arterial cisplatin. The anatomic structures whose malfunction was likely to cause each of the VF abnormalities common to both regimens were determined by literature review. Pre- and posttherapy CT scans were reviewed for evidence of posttherapy damage to each of these structures, and those demonstrating posttherapy changes were deemed dysphagia/aspiration-related structures (DARS). Standard three-dimensional (3D) RT, standard IMRT (stIMRT), and dysphagia-optimized IMRT (doIMRT) plans in which sparing of the DARS was included in the optimization cost function, were produced for each of 20 consecutive patients with advanced head-and-neck cancer.

RESULTS

The posttherapy VF abnormalities common to both regimens included weakness of the posterior motion of the base of tongue, prolonged pharyngeal transit time, lack of coordination between the swallowing phases, reduced elevation of the larynx, and reduced laryngeal closure and epiglottic inversion, contributing to a high rate of aspiration. The anatomic structures whose malfunction was the likely cause of each of these abnormalities, and that also demonstrated anatomic changes after RT concurrent with gemcitabine doses associated with dysphagia and aspiration, were the pharyngeal constrictor muscles (median thickness near midline 2.5 mm before therapy vs. 7 mm after therapy; p = 0.001), the supraglottic larynx (median thickness, 2 mm before therapy vs. 4 mm after therapy; p < 0.001), and, similarly, the glottic larynx. The constrictors and the glottic and supraglottic larynx were, therefore, deemed the DARS. The lowest maximal dose delivered to a stricture volume was 50 Gy. Reducing the volumes of the DARS receiving > or =50 Gy (V(50)) was, therefore, a planning and evaluation goal. Compared with the 3D plans, stIMRT reduced the V(50) of the pharyngeal constrictors by 10% on average (range, 0-36%, p < 0.001), and doIMRT reduced these volumes further, by an additional 10% on average (range, 0-38%; p <0.001). The V(50) of the larynx (glottic + supraglottic) was reduced marginally by stIMRT compared with 3D (by 7% on average, range, 0-56%; p = 0.054), and doIMRT reduced these volumes by an additional 11%, on average (range, 0-41%; p = 0.002). doIMRT reduced laryngeal V(50) compared with 3D, by 18% on average (range 0-61%; p = 0.001). Certain target delineation rules facilitated sparing of the DARS by IMRT. The maximal DARS doses were not reduced by IMRT because of their partial overlap with the targets. stIMRT and doIMRT did not differ in target doses, parotid gland mean dose, spinal cord, or nonspecified tissue maximal dose.

CONCLUSIONS

The structures whose damage may cause dysphagia and aspiration after intensive chemotherapy and RT are the pharyngeal constrictors and the glottic and supraglottic larynx. Compared with 3D-RT, moderate sparing of these structures was achieved by stIMRT, and an additional benefit, whose extent varied among the patients, was gained by doIMRT, without compromising target doses. Clinical validation is required to determine whether the dosimetric gains are translated into clinical ones.

A new approach in dose measurement and error analysis for narrow photon beams (beamlets) shaped by different multileaf collimators using a small detector

Dose measurement for narrow stereotactic beams and intensity modulation radiotherapy beamlets is difficult and error-prone due to the lack of lateral electron equilibrium. A small detector position error and finite sensitive volume as well as the nonfocus collimation could result in considerable (> 10%) measurement errors. A new method is introduced here to measure the dose and error components so that the accuracy and precision of the dose measurement can be improved. Based on superposition principle, we can create exactly the small field of interest by subtraction of a reference open field (O-field) and two strip fields (S-fields) from the sum of four quadrant fields (Q-fields). The position effect on the dose measurement is determined by the standard deviation of the four Q-field readings. The collimator leaf-edge effect (LEE) is quantified by the difference between the readings of the two S-fields using a detector that has very small sensitive volume. The detector-volume effect can be analytically estimated from the integrals of the dose distributions of the two S-fields over the detector volume. Using a pinpoint ion chamber (PTW N31006) and a stereotactic silicon diode detector (Scanditronix, DEB050), we have measured scatter factors (SF) and tissue-maximum ratios for 6-MV x-ray fields with sizes of 3 x 3 and 6 x 6 mm2 shaped by a BrainLAB micromultileaf collimator (microMLC) (M3), 4.4 x 4.4 and 8.8 x 8.8 mm2 shaped by a 3DLine double-focus MLC, and 5 x 5 and 10 x 10 mm2 by a Varian Millennium MLC. Our experimental results demonstrate that the large errors are often caused by a small setup error or measuring point displacement from the central ray of the beam. The LEE is almost independent of the depth but closely related to the field size and the type of MLC. The volume effect becomes significant when the detector diameter is comparable to the half size of the small fields. Application of the new method using different detectors had achieved less than 8.3% total experiment error for all of the small fields of interest except for the SF of the 3 x 3 mm2 field from the pinpoint ion chamber that has 15% volume effect. Importantly, the new method using a solid water phantom is clinically convenient and highly reproducible.

Prediction of respiratory tumour motion for real-time image-guided radiotherapy

Image guidance in radiotherapy and extracranial radiosurgery offers the potential for precise radiation dose delivery to a moving tumour. Recent work has demonstrated how to locate and track the position of a tumour in real-time using diagnostic x-ray imaging to find implanted radio-opaque markers. However, the delivery of a treatment plan through gating or beam tracking requires adequate consideration of treatment system latencies, including image acquisition, image processing, communication delays, control system processing, inductance within the motor, mechanical damping, etc. Furthermore, the imaging dose given over long radiosurgery procedures or multiple radiotherapy fractions may not be insignificant, which means that we must reduce the sampling rate of the imaging system. This study evaluates various predictive models for reducing tumour localization errors when a real-time tumour-tracking system targets a moving tumour at a slow imaging rate and with large system latencies. We consider 14 lung tumour cases where the peak-to-peak motion is greater than 8 mm, and compare the localization error using linear prediction, neural network prediction and Kalman filtering, against a system which uses no prediction. To evaluate prediction accuracy for use in beam tracking, we compute the root mean squared error between predicted and actual 3D motion. We found that by using prediction, root mean squared error is improved for all latencies and all imaging rates evaluated. To evaluate prediction accuracy for use in gated treatment, we present a new metric that compares a gating control signal based on predicted motion against the best possible gating control signal. We found that using prediction improves gated treatment accuracy for systems that have latencies of 200 ms or greater, and for systems that have imaging rates of 10 Hz or slower.

A new algorithm for determining collimator angles that favor efficiency in MLC based IMRT delivery

A new algorithm to determine collimator angles that favor delivery efficiency of intensity modulated radiotherapy plans was developed. It was found that the number of segments and monitor units (MUs) were largely reduced with the set of collimator angles determined with the new algorithm without compromising plan quality. The improvement of delivery efficiency using the new algorithm depends on the size and shape of the target(s), the number of modulation levels, and the type of leaf-sequencing algorithm. In a typical prostate case, when a sweeping leaf-sequencer is used for Varian 120 leaf (0.5 x 0.5 cm2 beamlet), 80 leaf (1 x 1 cm2 beamlet) and Elekta 40 leaf (1 x 1 cm2 beamlet), the number of segments was reduced by 42%, 29%, and 5%, respectively. The number of MUs was reduced by 41%, 35%, and 10%. For the Siemens MLC (IMFAST leaf sequencer, 1 x 1 cm2 beamlet) the segment reduction was 32% and the MU reduction was 14%. Comparison of the plans using the new and Brahme algorithms, in terms of target conformity index and dose volume histogram of the organs at risk, showed that the quality of the plans using the new algorithm was uncompromised. Similar results were obtained for a set of head and neck treatment plans.

Surface buildup dose dependence on photon field delivery technique for IMRT

The more complex delivery techniques required for implementation of intensity‐modulated radiotherapy (IMRT) based on inverse planning optimization have changed the relationship between dose at depth and dose at buildup regions near the surface. Surface buildup dose is dependent on electron contamination primarily from the unblocked view of the flattening filter and secondarily from air and collimation systems. To evaluate the impact of beam segmentation on buildup dose, measurements were performed with 10×10 cm2 fields, which were delivered with 3 static 3.5×10 cm2 or 3×10 cm2 strips, 5 static 2×10 cm2 strips, 10 static 1×10 cm2 strips, and 1.1×10 cm2 dynamic delivery, compared with a 10×10 cm2 open field. Measurements were performed in water and Solid Water using parallel plate chambers, a stereotactic diode, and thermoluminescent dosimeters (TLDs) for a 6 MV X‐ray beam. Depth doses at 2 mm depth (relative to dose at 10 cm depth) were lower by 6%, 7%, 11%, and 10% for the above field delivery techniques, respectively, compared to the open field. These differences are most influenced by differences in multileaf collimator (MLC) transmission contributing to the useful beam. An example IMRT field was also studied to assess variations due to delivery technique (static vs. dynamic) and intensity level. Buildup dose is weakly dependent on the multileaf delivery technique for efficient IMRT fields.

PACS numbers: 87.53.‐j, 87.53.Dq

Verification of step-and-shoot IMRT delivery using a fast video-based electronic portal imaging device

We present an investigation into the use of a fast video-based electronic portal-imaging device (EPID) to study intensity modulated radiation therapy (IMRT) delivery. The aim of this study is to test the feasibility of using an EPID system to independently measure the orchestration of collimator leaf motion and beam fluence; simultaneously measuring both the delivered field fluence and shape as it exits the accelerator head during IMRT delivery. A fast EPID that consists of a terbium-doped gadolinium oxysulphide (GdO2S:Tb) scintillator coupled with an inexpensive commercial 30 frames-per-second (FPS) CCD-video recorder (16.7 ms shutter time) was employed for imaging IMRT delivery. The measurements were performed on a Varian 2100 C/D linear accelerator equipped with a 120-leaf multileaf-collimator (MLC). A characterization of the EPID was performed that included measurements of spatial resolution, linac pulse-rate dependence, linear output response, signal uniformity, and imaging artifacts. The average pixel intensity for fields imaged with the EPID was found to be linear in the delivered monitor units of static non-IMRT fields between 3x3 and 15x15 cm2. A systematic increase of the average pixel intensity was observed with increasing field size, leading to a maximum variation of 8%. Deliveries of a clinical step-and-shoot mode leaf sequence were imaged at 600 MU/min. Measurements from this IMRT delivery were compared with experimentally validated MLC controller log files and were found to agree to within 5%. An analysis of the EPID image data allowed identification of three types of errors: (1) 5 out of 35 segments were undelivered; (2) redistributing all of the delivered segment MUs; and (3) leaf movement during segment delivery. Measurements with the EPID at lower dose rates showed poor agreement with log files due to an aliasing artifact. The study was extended to use a high-speed camera (1-1000 FPS and 10 micros shutter time) with our EPID to image the same delivery to demonstrate the feasibility of imaging without aliasing artifacts. High-speed imaging was shown to be a promising direction toward validating IMRT deliveries with reasonable image resolution and noise.