Differences In Output Factors Research Articles

A generalized model for monitor units determination in ocular proton therapy using machine learning: A proof-of-concept study

Objective. Determining and verifying the number of monitor units is crucial to achieving the desired dose distribution in radiotherapy and maintaining treatment efficacy. However, current commercial treatment planning system(s) dedicated to ocular passive eyelines in proton therapy do not provide the number of monitor units for patient-specific plan delivery. Performing specific pre-treatment field measurements, which is time and resource consuming, is usually gold-standard practice. This proof-of-concept study reports on the development of a multi-institutional-based generalized model for monitor units determination in proton therapy for eye melanoma treatments. Approach. To cope with the small number of patients being treated in proton centers, three European institutes participated in this study. Measurements data were collected to address output factor differences across the institutes, especially as function of field size, spread-out Bragg peak modulation width, residual range, and air gap. A generic model for monitor units prediction using a large number of 3748 patients and broad diversity in tumor patterns, was evaluated using six popular machine learning algorithms: (i) decision tree; (ii) random forest, (iii) extra trees, (iv) K-nearest neighbors, (v) gradient boosting, and (vi) the support vector regression. Features used as inputs into each machine learning pipeline were: Spread-out Bragg peak width, range, air gap, fraction and calibration doses. Performance measure was scored using the mean absolute error, which was the difference between predicted and real monitor units, as collected from institutional gold-standard methods. Main results. Predictions across algorithms were accurate within 3% uncertainty for up to 85.2% of the plans and within 10% uncertainty for up to 98.6% of the plans with the extra trees algorithm. Significance. A proof-of-concept of using machine learning-based generic monitor units determination in ocular proton therapy has been demonstrated. This could trigger the development of an independent monitor units calculation tool for clinical use.

Initial experiments with gel-water: towards MRI-linac dosimetry and imaging.

Tracking the position of a moving radiation detector in time and space during data acquisition can replicate 4D image-guided radiotherapy (4DIGRT). Magnetic resonance imaging (MRI)-linacs need MRI-visible detectors to achieve this, however, imaging solid phantoms is an issue. Hence, gel-water, a material that provides signal for MRI-visibility, and which will in future work, replace solid water for an MRI-linac 4DIGRT quality assurance tool, is discussed. MR and CT images of gel-water were acquired for visualisation and electron density verification. Characterisation of gel-water at 0 T was compared to Gammex-RMI solid water, using MagicPlate-512 (M512) and RMI Attix chamber; this included percentage depth dose, tissue-phantom ratio (TPR20/10), tissue-maximum ratio (TMR), profiles, output factors, and a gamma analysis to investigate field penumbral differences. MR images of a non-powered detector in gel-water demonstrated detector visualisation. The CT-determined gel-water electron density agreed with the calculated value of 1.01. Gel-water depth dose data demonstrated a maximum deviation of 0.7% from solid water for M512 and 2.4% for the Attix chamber, and by 2.1% for TPR20/10 and 1.0% for TMR. FWHM and output factor differences between materials were ≤0.3 and ≤1.4%. M512 data passed gamma analysis with 100% within 2%, 2mm tolerance for multileaf collimator defined fields. Gel-water was shown to be tissue-equivalent for dosimetry and a feasible option to replace solid water.

SU-F-T-610: Comparison of Output Factors for Small Radiation Fields Used in SBRT Treatment

Purpose: In order to fundamentally understand our previous dose verification results between measurements and calculations from treatment planning system (TPS) for SBRT plans for different sized targets, the goal of the present work was to compare output factors for small fields measured using EDR2 films with TPS and Monet Carlo (MC) simulations. Methods: 6MV beam was delivered to EDR2 films for each of the following field sizes; 1×1 cm2, 1.5×1.5 cm2, 2×2 cm2, 3×3 cm2, 4×4 cm2, 5×5 cm2 and 10×10 cm2. The films were developed in a film processer, then scanned with a Vidar VXR-16 scanner and analyzed using RIT113 version 6.1. A standard calibration curve was obtained with the 6MV beam and was used to get absolute dose for measured field sizes. Similar plans for all fields sizes mentioned above were generated using Eclipse with the Analytical Anisotropic Algorithm. Similarly, MC simulations were carried out using the MCSIM, an in-house MC code for different field sizes. Output factors normalized to 10×10 cm2 reference field were calculated for different field sizes in all the three cases and compared. Results: For field sizes ranging from 1×1 cm2 to 2×2 cm2, the differences in output factors between measurements (films), TPS and MC simulations were within 0.22%. For field sizes ranging from 3×3cm2 to 5×5cm2, differences in output factors were within 0.10%. Conclusion: No clinically significant difference was obtained in output factors for different field sizes acquired from films, TPS and MC simulations. Our results showed that the output factors are predicted accurately from TPS when compared to the actual measurements and superior dose calculation Monte Carlo method. This study would help us in understanding our previously obtained dose verification results for small fields used in the SBRT treatment.

SU‐F‐E‐06: Dosimetric Characterization of Small Photons Beams of a Novel Linear Accelerator

Purpose:The aim of the present contribution was to measure the main dosimetric quantities of small fields produced by UNIQUE and evaluate its matching with the corresponding dosimetric data of one 21EX conventional linear accelerator (Varian) in operation at the same center. The second step was to evaluate comparative performance of the EDGE diode detector and the PinPoint micro‐ionization chamber for dosimetry of small fields.Methods:UNIQUE is configured with MLC (120 leaves with 0.5 cm leaf width) and a single low photon energy of 6 MV. Beam data were measured with scanning EDGE diode detector (volume of 0.019 mm3), a PinPoint micro‐ionization chamber (PTW) and for larger fields (≥ 4×4cm2) a PTW Semi flex chamber (0.125 cm3) was used. The scanning system used was the 3D cylindrical tank manufactured by Sun Nuclear, Inc. The measurement of PDD and profiles were done at 100 cm SSD and 1.5 depth; the relative output factors were measured at 10 cm depth.Results:PDD and the profile data showed less than 1% variation between the two linear accelerators for fields size between 2×2 cm2 and 5×5cm2. Output factor differences was less than 1% for field sizes between 3×3 cm2 and 10×10 cm2 and less of 1.5 % for fields of 1.5×1.5 cm2 and 2×2 cm2 respectively. The dmax value of the EDGE diode detector, measured from the PDD, was 8.347 mm for 0.5×0,5cm2 for UNIQUE. The performance of EDGE diode detector was comparable for all measurements in small fields.Conclusion:UNIQUE linear accelerator show similar dosimetrics characteristics as conventional 21EX Varian linear accelerator for small, medium and large field sizes.EDGE detector show good performance by measuring dosimetrics quantities in small fields typically used in IMRT and radiosurgery treatments.

Monte Carlo calculated and experimentally determined output correction factors for small field detectors in Leksell Gamma Knife Perfexion beams

The measurement of output factors (OF) for the small photon beams generated by Leksell Gamma Knife® (LGK) radiotherapy units is a challenge for the physicist due to the under or over estimation of these factors by a vast majority of the detectors commercially available. Output correction factors, introduced in the international formalism published by Alfonso (2008 Med. Phys. 35 5179–86), standardize the determination of OFs for small photon beams by correcting detector-reading ratios to yield OFs in terms of absorbed-dose ratios. In this work output correction factors for a number of detectors have been determined for LGK Perfexion™ 60Co γ-ray beams by Monte Carlo (MC) calculations and measurements. The calculations were made with the MC system PENELOPE, scoring the energy deposited in the active volume of the detectors and in a small volume of water; the detectors simulated were two silicon diodes, one liquid ionization chamber (LIC), alanine and TLD. The calculated LIC output correction factors were within ± 0.4%, and this was selected as the reference detector for experimental determinations where output correction factors for twelve detectors were measured, normalizing their readings to those of the LIC. The MC-calculated and measured output correction factors for silicon diodes yielded corrections of up to 5% for the smallest LGK collimator size of 4 mm diameter. The air ionization chamber measurements led to extremely large output correction factors, caused by the well-known effect of partial volume averaging. The corrections were up to 7% for the natural diamond detector in the 4 mm collimator, also due to partial volume averaging, and decreased to within about ± 0.6% for the smaller synthetic diamond detector. The LIC, showing the smallest corrections, was used to investigate machine-to-machine output factor differences by performing measurements in four LGK units with different dose rates. These resulted in OFs within ± 0.6% and ± 0.2% for the 4 mm and 8 mm collimators, respectively, providing evidence for the use of generic OFs for these LGK beams. Using the experimentally derived output correction factors, OFs can be measured using a wide range of commercially available detectors.

SU‐E‐T‐52: Beam Data Comparison for 20 Linear Accelerators in One Network

Purpose:To compare photon beam data for the 20 Varian linear accelerators (TrueBeam, iX, and EX models) in use at five centers in the same network with the intent to model with one set of beam data in Eclipsec.Methods:Varian linear accelerators, TrueBeam (3), 21 EX, iX, and Trilogy (14), and 6 EX (3), installed between 1999 and 2014 have their 6 MV and 15 MV x‐ray beams reevaluated. Full commissioning, including output factors (St), percent depth doses (PDD), and off‐axis profiles, was recently performed for a TrueBeam with a cc04 ion chamber in an IBA Blue phantom. Similarly, a subset of beam data for each of the other accelerators was measured recently as follows: for 3×3, 10×10, and 30×30 cm2 field sizes, flatness and penumbra (80–20%) were measured at dmax and 10 cm depths, PDD were measured at 10 and 20 cm depths, and St were measured at 5 cm depth. Measurement results for all machines were compared.Results:For 15 high‐energy (6 and 15 MV) and 3 low‐energy machines (6MV only): 1) PDD agreed within 1.4% at 10 and 20 cm depths; 2) penumbra agreed within 1.0 mm at dmax and 10 cm depths; 3) flatness was within 1.3% at dmax and 10 cm depths; and 4) with exception of the three low energy machines, output factors were within 1.1% and 0.5% for 3×3 and 30×30 cm2, respectively. Measurement uncertainty, not quantified here, accounts for some of these differences.Conclusion:Measured beam data from 15 high‐energy Varian linacs are consistent enough that they can be classified using one beam data set in Eclipse. Two additional high‐energy machines are removed from this group until their data are further confirmed. Three low‐energy machines will be in a separate class based upon differences in output factors (St).

SU‐E‐T‐33: A Comparative Study of Commissioning Data of Two TrueBeam LINACs with TrueBeam Representative Beam Data

Purpose: To compare the commissioning water tank data of the two TrueBeam type linacs with TrueBeam Representative Beam Data. Methods: Three flattening filtered (FF), 6MV, 10MV, and 15MV, and two flattening filtered free (FFF), 6FFF and 10FFF, photon beams of one TrueBeam and one TrueBeam STX Linacs (Varian Medical Systems) are compared to the TrueBeam Representative Beam Data (TRBD) as provided by Varian. All measurements are scanned in Blue Phantom (IBA) with cc04 ion chamber at SSD 100cm at the highest dose rate available for each beam. The TRBD set is the average values of three TrueBeams at a single institution acquired by Iba cc13 ion chamber. To compare the beam quality, the PDD at d10 and d20 and the corresponding PDD 10/20 ratio at field sizes of 3×3, 10×10 and 30×30 cm2 are compared. The flatness of 10×10 and 30×30 cm2 open field profiles at depth of dmax, 10cm and 20cm are also compared. The output factors for square fields are also compared. Results: The PDD difference at both d10 and d20 between all FF and FFF beams and TRBD are within 0.6% and 0.9% respectively. The average PDD ratio difference between all beams and TRBD is within 0.5%. The flatness of all the FF and FFF beams are within 0.4% and 1.0% of TRBD respectively. The output factor differences between the two TrueBeams and RTBD for all beams and field sizes are within 0.8%. Conclusion: All the beam parameters of both the FF and FFF beams of the two Truebeams are within 1% of TRBD. Research grant from Varian Medical Systems

Commissioning measurements for photon beam data on three TrueBeam linear accelerators, and comparison with Trilogy and Clinac 2100 linear accelerators

This study presents the beam data measurement results from the commissioning of three TrueBeam linear accelerators. An additional evaluation of the measured beam data within the TrueBeam linear accelerators contrasted with two other linear accelerators from the same manufacturer (i.e., Clinac and Trilogy) was performed to identify and evaluate any differences in the beam characteristics between the machines and to evaluate the possibility of beam matching for standard photon energies. We performed a comparison of commissioned photon beam data for two standard photon energies (6 MV and 15 MV) and one flattening filter‐free (“FFF”) photon energy (10 FFF) between three different TrueBeam linear accelerators. An analysis of the beam data was then performed to evaluate the reproducibility of the results and the possibility of “beam matching” between the TrueBeam linear accelerators. Additionally, the data from the TrueBeam linear accelerator was compared with comparable data obtained from one Clinac and one Trilogy linear accelerator models produced by the same manufacturer to evaluate the possibility of “beam matching” between the TrueBeam linear accelerators and the previous models. The energies evaluated between the linear accelerator models are the 6 MV for low energy and the 15 MV for high energy. PDD and output factor data showed less than 1% variation and profile data showed variations within 1% or 2 mm between the three TrueBeam linear accelerators. PDD and profile data between the TrueBeam, the Clinac, and Trilogy linear accelerators were almost identical (less than 1% variation). Small variations were observed in the shape of the profile for 15 MV at shallow depths (< 5 cm) probably due to the differences in the flattening filter design. A difference in the penumbra shape was observed between the TrueBeam and the other linear accelerators; the TrueBeam data resulted in a slightly greater penumbra width. The diagonal scans demonstrated significant differences in the profile shapes at a distance greater than 20 cm from the central axis, and this was more notable for the 15 MV energy. Output factor differences were found primarily at the ends of the field size spectrum, with observed differences of less than 2% as compared to the other linear accelerators. The TrueBeam's output factor varied less as a function of field size than the output factors for the previous models; this was especially true for the 6 MV. Photon beam data were found to be reproducible between different TrueBeam linear accelerators well within the accepted clinical tolerance of ±2%. The results indicate reproducibility in the TrueBeam machine head construction and a potential for beam matching between these types of linear accelerators. Photon beam data (6 MV and 15 MV) from the Trilogy and Clinac 2100 showed several similarities and some small variations when compared to the same data measured on the TrueBeam linear accelerator. The differences found could affect small field data and also very large field sizes in beam matching considerations between the TrueBeam and previous linear accelerator models from the same manufacturer, but should be within the accepted clinical tolerance for standard field sizes and standard treatments.PACS number: 87.56. bd

Validation of the final aperture superposition technique to calculate electron output factors and depth dose curves

The final aperture superposition technique (FAST) is a method to reproduce rapidly the electron-beam depth dose curves and output factors that would be calculated by a full Monte Carlo simulation. FAST uses precalculated Monte Carlo-based differential dose arrays and performs a superposition of open and shielded contributions to account for arbitrarily shaped insert openings. The objective of this work was to refine and validate the accuracy of the FAST method for a full range of treatment parameters. Compared to full simulations, raw FAST calculations tended to underestimate dose near the surface deposited by particles that crossed the shield-opening interface of the insert. In this study, a set of empirical correction curves was derived to reduce the errors from this "collimator effect." FAST and full simulation calculations were compared for every combination of six beam energies (6-21 MeV), four applicator sizes (10-25 cm), and two source-to-surface distances (SSDs) (100 and 110 cm). Validation tests were performed for a total of 192 fields using four sample insert openings: an open insert and 2, 3, and 5 cm diameter circular openings. Calculations were also performed for four patient inserts with irregularly shaped openings. Using the empirical correction curves, systematic errors were reduced, resulting in mean dose differences of less than 1% of the maximum full simulation dose. FAST relative output factors reproduced full simulation output factors to within 3% for all configurations except for the 2 and 3 cm diameter openings for the 6 and 9 MeV beams at 110 cm SSD. The maximum shift between the FAST and full simulation depth dose curves in the 90%-80% fall-off region was less than 3 mm for 97% of the fields. For the patient insert calculations, differences in output factors and mean differences in depth dose curves were within 1.5% with maximum shifts of 1.5 mm in the 90%-80% fall-off region. A small set measurements also demonstrated 3% accuracy in FAST output factors except for a 5% deviation for a 2 cm diameter insert for the 6 MeV beam at 110 cm SSD. These results demonstrate that FAST can be used to provide output factors and depth dose curves for most clinical cases.

Accuracy of central axis dose calculations for photon external radiotherapy beams in Finland: The quality of local beam data and the use of averaged data

Purpose The accuracy of central axis dose calculation was evaluated for 48 photon beams from 28 linear accelerators at nine centres in Finland. In addition, inter-accelerator consistency of beam data was evaluated for Varian Clinac 600 CDs and 2100 CDs, and averaged data sets were generated for output factors (OFs) and percentage depth doses (PDDs). The averaged data sets obtained were used to identify potential dosimetry reasons for local errors. Materials and methods Agreement between measured and calculated doses was determined at isocentre at 10 cm depth in water for nine different sized open square and rectangular fields. Averaged OFs were determined for nominal energies of 4, 6, 10, 15 and 18 MV both at d max and at a 10-cm depth. In order to develop a function for the OF data, OFs for square fields were parametrised through empirical model fitting. The feasibility of a simple equivalent square collimator formula was also evaluated for the presentation of OFs for rectangular fields. Averaged PDDs were determined at a 10-cm depth. Results The difference between measured and calculated doses exceeded ±3%, ±2% and ±1% for 3, 6 and 35 of the investigated 48 beams, respectively. The differences were due to errors observed in both OFs and depth dose data. When the agreement between dose calculation and measurement was within ±1%, inter-accelerator differences in OFs were within ±1.0% at both the depth of dose maximum and at 10 cm for Clinac 600 CDs and also for 2100 CDs. Differences in PDDs were within ±1.2%. Conclusions The importance of quality control for beam data was demonstrated by showing significant errors in measured data. For Clinac 600 and 2100 CDs, the quality control can be accurately performed by comparing local data to averaged reference data. Robust averaged data sets were obtained for 6, 15 and 18 MV beams of Clinac 2100 CDs.

Basic dosimetric verification in water of the anisotropic analytical algorithm for Varian, Elekta and Siemens linacs

Since early 2007 a new version of the Anisotropic Analytical Algorithm (AAA) for photon dose calculations was released by Varian Medical Systems for clinical usage on Elekta linacs and also, with some restrictions, for Siemens linacs. Basic validation studies were performed and reported for three beams: 4,6 and 15 MV for an Elekta Synergy, 6 and 15 MV for a Siemens Primus and, as a reference, for 6 and 15 MV from a Varian Clinac 2100C/D. Generally AAA calculations reproduced well measured data and small deviations were observed for open and wedged fields. PDD curves showed in average differences between calculation and measurement smaller than 1% or 1.2 mm for Elekta beams, 1% or 1.8 mm for Siemens beams and 1% or 1 mm for Varian beams. Profiles in the flattened region matched measurements with deviations smaller than 1% for Elekta and Varian beams, 2% for Siemens. Percentage differences in Output Factors were observed as small as 1% in average.

Strahlenphysikalische Einflussgrößen bei der Dosimetrie mit verschiedenen Detektortypen

The present study investigated the radiophysical influences on the measurement of dosimetry basic data, attributable to field size, photon energy and detector type. A natural diamond detector, two ionisation chambers, different Si-diodes and a EBT-Gafchromic film were studied for this purpose. The characteristics of the detectors were investigated with regard to the measurement of output factors, lateral beam profiles and relative depth-dose curves for narrow and wide photon beams of 15 MV. Significant differences in output factors were obtained with different detectors. For narrow fields, the natural diamond detector and the diodes PTW-60012 and SCX_WH-PFD measured output factors close to those of the EBT-Gafchromic film. The output factor for large fields was overestimated by the unshielded diode PTW 60012 and the PinPoint-chamber PTW-31006 because of their over-response to scattered photons. The relative depth dose distributions for wide beams at large depths agree well for the diamond, the ionisation chambers and the shielded Diode SCX_WH-PFD and PTW-60008, while the measured dose was overestimated by an unshielded diode PTW-60012. Considering the influence due to the sensitive materials and the construction of the detectors the manufacturers of dosimeters have specified the application ranges for the various types of detectors.

Dosimetric validation of the anisotropic analytical algorithm for photon dose calculation: fundamental characterization in water

In July 2005 a new algorithm was released by Varian Medical Systems for the Eclipse planning system and installed in our institute. It is the anisotropic analytical algorithm (AAA) for photon dose calculations, a convolution/superposition model for the first time implemented in a Varian planning system. It was therefore necessary to perform validation studies at different levels with a wide investigation approach. To validate the basic performances of the AAA, a detailed analysis of data computed by the AAA configuration algorithm was carried out and data were compared against measurements. To better appraise the performance of AAA and the capability of its configuration to tailor machine-specific characteristics, data obtained from the pencil beam convolution (PBC) algorithm implemented in Eclipse were also added in the comparison. Since the purpose of the paper is to address the basic performances of the AAA and of its configuration procedures, only data relative to measurements in water will be reported. Validation was carried out for three beams: 6 MV and 15 MV from a Clinac 2100C/D and 6 MV from a Clinac 6EX. Generally AAA calculations reproduced very well measured data, and small deviations were observed, on average, for all the quantities investigated for open and wedged fields. In particular, percentage depth–dose curves showed on average differences between calculation and measurement smaller than 1% or 1 mm, and computed profiles in the flattened region matched measurements with deviations smaller than 1% for all beams, field sizes, depths and wedges. Percentage differences in output factors were observed as small as 1% on average (with a range smaller than ±2%) for all conditions. Additional tests were carried out for enhanced dynamic wedges with results comparable to previous results. The basic dosimetric validation of the AAA was therefore considered satisfactory.