Dosimetry and quality specification of high energy photon beams

Estimation of the lateral variation of photon beam energy spectra using the percentage depth dose reconstruction method

In photon-collapsed cone convolution (pCCC) algorithm of the Monaco treatment planning system (TPS), the central-axis energy spectrum is assumed constant throughout the entire irradiation area. To consider lateral variations, an off-axis softening factor is applied to attenuation coefficients during the total energy released per unit mass calculation. We evaluated this method through comparison studies of percentage depth doses (PDDs) and off-axis ratios (OARs) calculated by Monaco and measured for a 6 MV photon beam at various off-axis angles and depths. Significant differences were observed, with relative differences exceeding ± 1%. Therefore, this method may not accurately represent lateral variations of energy spectra. We propose directly implementing energy spectra on both central-axis and off-axis to improve dose calculation accuracy for large field. To this end, we introduce reconstruction of PDDs from monoenergetic depth doses (MDDs) along off-axis angles, thereby estimating energy spectra as functions of radial distance. This method derives energy spectra quickly without significantly increasing the beam modeling time. MDDs were computed through Monte Carlo simulations (DOSRZnrc). The variances between reconstructed and measured PDDs were minimized using the generalized-reduced-gradient method to optimize energy spectra. Reconstructed PDDs along off-axis angles of 0°, 1.15°, 2.29°, 3.43°, 4.57°, 5.71°, 6.84°, 7.97°, 9.09°, 10.2° to estimate energy spectra at radial distances of 0-18 cm in 2 cm increments and OARs calculated using estimated energy spectra at 5, 10, and 20 cm depths, well agreed with measurement (relative differences within ± 0.5%). In conclusion, our proposed method accurately estimates lateral energy spectrum variation, thereby improving dose calculation accuracy of pCCC algorithm.

Analysis of Uncertainties in Clinical High-Energy Photon Beam Calibrations Using Absorbed Dose Standards

We compared the results of absorbed dose measurements made using the TRS-398, TG-51, and DIN protocols and their associated uncertainties to reduce discrepancies in measurement results made using the three protocols. This experiment was carried out on two Varian Medical linear accelerators with 4, 6, 10, and 20 MV photon energies using FC65-G and CC15 (cylindrical) and NACP-02-type (plane-parallel) ion chambers in water phantoms. The radiation beam quality index (Q) was determined from the measurement of percentage depth dose. It was used to determine the photon beam quality factor required with the ionization chamber calibration factor to convert the ion chamber reading into the absorbed dose to water. For the same beam quality, the TRS-398/TG-51 varied from 0.01% to 1.8%, whereas the ratio for TRS-398/DIN 6800-2 varied from 0.1% to 0.88%. The chamber-to-chamber variation was 0.09% in TRS-398/TG-51, 0.03% in TRS-398/DIN, and 0.02% in TG-51/DIN 6800-2. The expanded uncertainties (k = 1) were 1.24 and 1.25 when using TRS-398 and DIN 6800-2, respectively. Using the aforementioned three protocols, the results showed little chamber-to-chamber variation and uncertainty in absorbed dose measurements. The estimated uncertainties when using cylindrical ion chambers were slightly lower than those measured using plane-parallel chambers. The results are important in facilitating comparisons of absorbed dose measurements when using the three protocols.

Artificial intelligence in radiotherapy: a technological review

Radiation therapy (RT) is widely used to treat cancer. Technological advances in RT have occurred in the past 30 years. These advances, such as three-dimensional image guidance, intensity modulation, and robotics, created challenges and opportunities for the next breakthrough, in which artificial intelligence (AI) will possibly play important roles. AI will replace certain repetitive and labor-intensive tasks and improve the accuracy and consistency of others, particularly those with increased complexity because of technological advances. The improvement in efficiency and consistency is important to manage the increasing cancer patient burden to the society. Furthermore, AI may provide new functionalities that facilitate satisfactory RT. The functionalities include superior images for real-time intervention and adaptive and personalized RT. AI may effectively synthesize and analyze big data for such purposes. This review describes the RT workflow and identifies areas, including imaging, treatment planning, quality assurance, and outcome prediction, that benefit from AI. This review primarily focuses on deep-learning techniques, although conventional machine-learning techniques are also mentioned.

Dosimetric verification of compensated beams using radiographic film

INTRODUCTION

External photon beam modulation using compensators in order to achieve a desired dose distribution when brachytherapy treatment is followed by external beam radiation is a well-established technique. A compensator modulates the central part of the beam, and the dose beneath the thickest part of the compensator is delivered mostly by scattered, low energy photons. A two-dimensional detector with a good spatial resolution is needed for the verification of those beams. In this work, the influence of different types of detectors on the measured modulated dose distributions was examined.

MATERIALS AND METHODS

Dosimetric verification was performed using X-Omat V, Eastman Kodak radiographic films at different depths in a solid water phantom. The film measurements were compared with those made by ionization chambers. Photon beams were also modelled using EGSnrc Monte Carlo algorithm to explain the measured results.

RESULTS

Monte Carlo calculated over-response of the film under the thickest part of the compensator was over 15%, which was confirmed by measurements. The magnitude of over-response could be associated with changes in the spectra of photon energy in the beam.

CONCLUSIONS

The radiographic film can be used for the dosimetry of compensated high energy photon beams, with limitations in volumes where photon spectra are hardly degraded.

Prediction of stopping-power ratios in flattening-filter free beams

PURPOSE

In recent years, there has been an increasing interest in flattening-filter free (FFF) beams. However, since the removal of the flattening filter will affect both the mean and the variance of the energy spectrum, current beam-quality specifiers may not be adequate for reference dosimetry in such beams. The purpose of this work was to investigate an alternative, more general beam-quality specifier.

METHODS

The beam-quality specifier used in this work was a combination of the kerma-weighted mean and the coefficient of variation of the linear attenuation coefficient in water. These parameters can in theory be determined from narrow-beam transmission measurements using a miniphantom "in-air," which is a measurement condition well suited also to small and nonstandard fields. The relation between the Spencer-Attix stopping-power ratios and this novel beam-quality specifier was described by a simple polynomial. For reference, the authors used Monte Carlo calculated spectra and stopping-power data for nine different beams, with and without flattening filter.

RESULTS

The polynomial coefficients were obtained by least-squares optimization. For all beams included in this investigation, the average of the differences between the predicted and the Monte Carlo calculated stopping-power ratios was 0.02 +/- 0.17% (1 SD) (including TomoTherapy and CyberKnife example beams).

CONCLUSIONS

An alternative dual-parameter beam-quality specifier was investigated. The evaluation suggests that it can be used successfully to predict stopping-power ratios in FFF as well as conventional beams, regardless of filtration.

Comparison between TG-51 and TRS-398: electron contamination effect on photon beam-quality specification

Two dosimetry protocols based on absorbed dose to water have recently been implemented: TG-51 and TRS-398. These protocols use different beam-quality indices: %dd(10)x and TPR20,10. The effect of electron contamination in measurements of %dd(10)x has been proposed as a disadvantage of the TG-51. For actual measurements of %dd(10)x in five clinical beams (Primus 6-18 MV, SL-75/5 6 MV, SL-18 6-15 MV) a purging magnet was employed to remove the electron contamination. Also, %dd(10)x was measured in the different ways described in TG-51 for high-energy beams: with a lead foil at 50 cm from the phantom surface, at 30 cm, and for open beam. Moreover, TPR20,10 was determined. Also, periodic quality-control measurements were used for comparing both quality indices and variation over time, but D20,10 was used instead of TPR20,10 and measurements in open beam for the %dd(10)x determination. Considering both protocols, S(w,air) and kQ were calculated in order to compare the results with the experimental data. Significant differences (0.3% for kQ) were only found for the two high-energy beams, but when the electron contamination is underestimated by TG-51, the difference in kQ is lower. Differences in the other cases and variations over time were less than 0.1%.

On beam quality and stopping power ratios for high-energy x-rays

The aim of this work is to quantitatively compare two commonly used beam quality indices, IPR(20/10) and %dd(10)x, with respect to their ability to predict stopping power ratios (water to air), s(w,air), for high-energy x-rays. In particular, effects due to a varied amount of filtration of the photon beam will be studied. A new method for characterizing beam quality is also presented, where the information we strive to obtain is the moments of the spectral distribution. We will show how the moments enter into a general description of the transmission curve and that it is possible to correlate the moments to s(w,air) with a unique and simple relationship. Comparisons with TPR(20/10) and %dd(10), show that the moments are well suited for beam quality specification in terms of choosing the correct s(w,air).

Composite energy deposition kernels for focused point monodirectional photon beams

A 3D volume overlap algorithm has been developed for converting energy deposition kernels between arbitrary 2D and 3D irradiation geometries. The kernels can be used as convolution kernels in inverse radiation therapy planning and as accurate descriptions of the dose distributions for clinically important beam geometries. The new method of dose calculation combining Monte Carlo and analytical methods has introduced an improved accuracy in dose calculation on the fractional per cent level. The comparisons are also made for a wide range of photon spectra and irradiation geometries from narrow point monodirectional pencil beams to finite uniform beams, 4pi steradians isotropically converging beams and divergent beams from isotropic point sources. It is seen that the photon scatter penumbra is highest at low photon energies whereas the secondary electron penumbra is widest at high photon energies, making low energy beams more interesting for small targets and high energy beams most useful for large deep-seated targets.

Depth for dose calibration in high energy photon beams

BACKGROUND AND PURPOSE

The normalisation depth for determination of output factors in photon fields has frequently been the depth of dose maximum. At high energies the contribution from contaminating electrons is significant at dose maximum and is critically dependent on the beam geometry parameters, which is why a larger depth should be preferred.

MATERIALS AND METHODS

The effect of electron contamination was studied using a purging magnet to remove charged particles from the treatment head and a helium bag to minimise production between the head and the phantom.

RESULTS

A depth of 10 cm was found to be beyond the range of the contaminating electrons for photon energies up to 20 MV (TPR(20)(10) = 0.772). However, at 50 MV (TPR(20)(10) = 0.810) contaminating electrons contribute 2-3% to the absorbed dose at 10 cm depth.

CONCLUSIONS

10 cm is recommended as both reference and normalisation depth for all megavoltage photon beam qualities, i.e. 60Co and X-rays from accelerators up to 50 MV.

The role of phantom and treatment head generated bremsstrahlung in high-energy electron beam dosimetry

An analytical expression has been derived for the phantom generated bremsstrahlung photons in plane-parallel monoenergetic electron beams normally incident on material of any atomic number (Be, H2O, Al, Cu and U). The expression is suitable for the energy range from 1 to 50 MeV and it is solely based on known scattering power and radiative and collision stopping power data for the material at the incident electron energy. The depth dose distribution due to the bremsstrahlung generated by the electrons in the phantom is derived by convolving the bremsstrahlung energy fluence produced in the phantom with a simple analytical energy deposition kernel. The kernel accounts for both electrons and photons set in motion by the bremsstrahlung photons. The energy loss by the primary electrons, the build-up of the electron fluence and the generation, attenuation and absorption of bremsstrahlung photons are all taken into account in the analytical formula. The longitudinal energy deposition kernel is derived analytically and it is consistent with both the classical biexponential relation describing the photon depth dose distribution and the exponential attenuation of the primary photons. For comparison Monte Carlo calculated energy deposition distributions using ITS3 code were used. Good agreement was found between the results with the analytical expression and the Monte Carlo calculation. For tissue equivalent materials, the maximum total energy deposition differs by less than 0.2% from Monte Carlo calculated dose distributions. The result can be used to estimate the depth dependence of phantom generated bremsstrahlung in different materials in therapeutic electron beams and the bremsstrahlung production in different electron absorbers such as scattering foils, transmission monitors and photon and electron collimators. By subtracting the phantom generated bremsstrahlung from the total bremsstrahlung background the photon contamination generated in the treatment head can be determined to allow accurate dosimetry of heavily photon contaminated electron beams.

Characteristics of contamination electrons in high energy photon beams

A simple method to estimate the contribution of contaminating electrons to the dose, and to evaluate their dosimetric characteristics is proposed. The method is based on a normalisation of the tissue--maximum ratio curves to a constant primary photon fluence. The contribution of the contaminating electrons to the dose is calculated by subtracting the dose relative to a small field from the dose relative to the field under consideration. The method includes the determination of the mean energy, the linear apparent attenuation coefficient, the 50% range and the maximum range of the contaminating electrons. The extrapolated surface dose normalised to a constant primary photon fluence has been found to be constant for a constant collimator opening whatever may be the source distance.

The status of high-energy photon and electron beam dosimetry five years after the implementation of the IAEA Code of Practice in the Nordic countries

The status of the dosimetry of high-energy photon and electron beams is analysed, taking into account the main developments in the field since the implementation of the IAEA Code of Practice in the Nordic countries. In electron beam dosimetry, energy-range relationships are discussed; Monte-Carlo results with different codes are compared with the experimentally derived empirical expression used in most protocols. Updated calculations of water-to-air stopping-power ratios following the changes in the Monte-Carlo code used to compute actual Sw,air values are compared with the data included in most dosimetry protocols. The validity of the commonly used procedure to select stopping-power ratios for a clinical beam from the mean energy at the phantom surface and the depth of measurement, is analysed for 'realistic' electron beams. In photon beam dosimetry, calculated correction factors including the effect of the wall plus waterproofing sleeve and existing data on the shift of the effective point of measurement of an ionization chamber, are discussed. New calculations of medium-to-air stopping-power ratios and their correlation with the quality of the beam obtained from the convolution of Monte-Carlo kernels are presented together with their possible practical implications in dosimetry. Trends in Primary Standard Dosimetry Laboratories towards implementing calibrations in terms of absorbed dose to water are presented, emphasizing controversial proposals for the specification of photon beam qualities. Plane-parallel ionization chambers are discussed regarding aspects that affect determinations of absorbed dose, either through the different methods used for the calibration of these chambers or by means of correction factors. Recent studies on the effect of the central electrode in Farmer-type cylindrical chambers are described.

Uncertainties in dosimetric data and beam calibration

Recent studies indicate that the calibration of therapeutic beams is one of the main sources of uncertainty in the mean absorbed dose to the target volume in radiotherapy. Interaction coefficients and data used through the different steps in the calibration are pointed out as the main contribution to this uncertainty. Procedures used to select dosimetric data, that is, input parameters used in the specification of the quality of the beam, cause another contribution. In this paper the actual status of the data used for the dosimetry of photon and electron beams is introduced first. Uncertainties along the dosimetric chain are analyzed according to the procedure and data used in recent publications. Uncertainties in stopping-power ratios, considered the main contribution, are discussed in detail starting from the basic electron stopping-power data. Overall uncertainties in the presently available set of stopping-power ratios are analyzed. Recent developments in the dosimetry of electron beams, related to the effect of energy and angular spread and electron and photon contamination, are discussed in connection with the procedure to select stopping-power ratios for clinical dosimetry. Uncertainties along the dosimetric chain are evaluated in terms of the present knowledge of error sources.

Radiation qualities of x-ray beams in cooperative clinical trials

X-ray beams are usually described by "MV" numbers which represent accelerating potentials (AP) and approximations to the maximum energies in the photon spectra. However, these numbers do not uniquely specify the properties of the beams. Current high energy photon dosimetry protocols specify radiation quality in terms of a measured ionization ratio which is equivalent to the ratio of the tissue-maximum ratios at depths 10 cm and 20 cm, for field size 10 cm X 10 cm [TMR)20(10]. For convenience, the American Association of Physicists in Medicine introduced a new parameter, known as the Nominal Accelerating Potential (NAP), which was derived from (TMR)20(10) and features values in MV units that are similar to those of the conventional accelerating potentials. (TMR)20(10) and Nominal Accelerating Potential may be considered to be expressions of the penetrating powers of x-ray beams. We determined (TMR)20(10) and Nominal Accelerating Potential for 460 treatment machines with stated accelerating potentials from 4 MV to 25 MV in the Quality Assurance Review Center's files of machine data from institutions that participate in cooperative clinical trials. The results demonstrate appreciable variability of the two parameters at each stated accelerating potential, with overlapping of adjacent groups of machines. It is concluded that the manufacturers' MV numbers do not reliably identify x-ray beams in terms of their depth dose properties. To promote standardization and consistency of energy specification in clinical trials as well as in general practice, we propose that x-ray beams be designated by their Nominal Accelerating Potential values as an adjunct to the use of (TMR)20(10) in radiation therapy.

Determination of effective bremsstrahlung spectra and electron contamination for photon dose calculations

A method is described for determining an effective, depth dose consistent bremsstrahlung spectra for high-energy photon beams using depth dose curves measured in water. A simple, analytical model with three parameters together with the nominal accelerating potential is used to characterise the bremsstrahlung spectra. The model is used to compute weights for depth dose curves from monoenergetic photons. These monoenergetic depth doses, calculated with the convolution method from Monte Carlo generated point spread functions (PSF), are added to yield the pure photon depth dose distribution. The parameters of the analytical spectrum model are determined using an iterative technique to minimise the difference between calculated and measured depth dose curves. The influence from contaminant electrons is determined from the difference between the calculated and the measured depth dose.

Design principles and clinical possibilities with a new generation of radiation therapy equipment. A review

The main steps in the development of isocentric megavoltage external beam radiation therapy machines are briefly reviewed identifying three principal types or generations of equipment to date. The new fourth generation of equipment presented here is characterized by considerably increased flexibility in dose delivery through the use of scanned elementary electron and photon beams of very high quality. Furthermore the wide energy range and the possibility of using high resolution multileaf collimation with all beam modalities makes it possible to simplify irradiation techniques and increase the accuracy in dose delivery. The main design features are described including a dual dipole magnet scanning system, a photon beam purging magnet, a helium atmosphere in the treatment head, a beam's eye view video read-out system of the collimator setting and a radiotherapeutic computed tomography facility. Some of the clinical applications of this new type of radiation therapy machine are finally reviewed, such as the ease of performing beam flattening, beam filtering and compensation, and the simplification of many treatment techniques using the wide spectrum of high quality electron and photon beams. Finally the interesting possibility of doing conformation and more general isocentric treatments with non-uniform beams using the multileaf collimator and the scanning system are demonstrated.

Calculation and application of point spread functions for treatment planning with high energy photon beams

A general dose calculation method for treatment planning with high energy photon beams, based on folding of the total energy released by primary photons per unit mass, the terma, with a fractional mean energy imparted point spread function is described. A set of point spread functions has been calculated with Monte Carlo technique for energies of primary photons between 100 keV and 20 MeV. Dose distributions have been calculated for a 6 MV bean using the method. The results clearly point out the considerably increased precision and flexibility achieved when calculating photon beam dose distributions from first principles using Monte Carlo generated point spread functions. The point spread functions calculated in this work are available on magnetic tape from the authors.

Radiotherapeutic computed tomography with scanned photon beams

Radiotherapeutic computed tomography is a powerful technique to generate anatomical transversal tomograms of the patient in treatment position by using the therapy beam from the treatment unit. For this purpose the treatment unit has to be equipped with a detector array that can detect the beam transmitted through the patient and a computer that analyzes the data and performs the back projection. When the treatment unit uses scanned elementary photon beams, the only practical technique available for generating high quality high energy photon beams, the operation principle and, to some extent, the image quality is similar to that of a 3rd generation CT-scanner. The optimum choice of detection geometry and type of radiation detectors for radiotherapeutic computed tomography particularly at high photon energies are discussed indicating the merits of BGO (bismuthgermanate) or CWO (cadmiumtungstate) photodiod arrays. The first tomographic images of a thorax phantom at an acceleration potential of 50 MV using such detectors are presented. The image contrast is similar to that for 300 kV X rays mainly because the considerable influence of pair production at 50 MV. Line spread and modulation transfer functions are presented indicating a resolution of the order of two millimeters using a crystal thickness of 5 mm. The advantages with radiotherapeutic computed tomography, beside forming a new general communication channel between different diagnostic techniques, dose planning, and radiation delivery, are the elimination of position errors and the provision of exact attenuation data for dose planning.