Detective quantum efficiency of an amorphous selenium detector to megavoltage radiation.

The spatial frequency dependent detective quantum efficiency (DQE(f)) of a high-resolution selenium-based imaging system has been measured at megavoltage energies. These results have been compared with theoretical calculations. The imaging system was a video tube with a 5 microm amorphous selenium (a-Se) target which was irradiated by 1.25 MeV gamma-rays. The modulation transfer function (MTF) decreased rapidly with spatial frequency (determined by spread of electrons in the build-up material) while the noise power spectrum was constant as a function of spatial frequency. The DQE obtained from these MTF and noise power measurements was compared with a Monte Carlo model of the pulse height spectrum of the detector. The DQE(0) model accounted for the interaction of x rays with the detector as well as the energy-dependent gain (charge generated/energy deposition). Good agreement between the calculated and measured DQE(0) was found. The model was also used to estimate the DQE(f) of a metal plate + a-Se detector which was compared with a metal plate + phosphor system of the same mass thickness. The DQE(f) s of both detectors are very similar, indicating that the choice of which detector is better will be based upon criteria other than DQE(f), such as read-out approach, ease of manufacture or sensitivity.

Sensitivity of amorphous selenium to x rays from 40 kVp to 18 MV: measurements and implications for portal imaging.

Recently, the clinical application of electronic portal imaging devices has enabled more frequent verification of patient setup for radiation treatment. However, the image quality has sometimes proven to be inadequate, motivating the investigation of alternative sensors with better image quality. Amorphous selenium (a-Se) is potentially one such sensor since the electrostatic image formation process has high resolution. To fully evaluate the potential of a-Se for portal imaging, it is necessary to investigate all the imaging properties at high x-ray energies. Here, measurements of the sensitivity of a-Se to incident x-ray spectra ranging in energy from 40 kVp to 18 MV and for a-Se thicknesses ranging from approximately 10 to 300 microns under full buildup conditions are described. When x rays or energetic electrons deposit energy in a photoconductor with an applied electric field, F, electrons and holes are released. The x-ray conversion sensitivity may be defined as 1/W +/-, where W +/- is the energy required to release an electron-hole pair. Consistent with the results of previous investigators, W +/- is found to vary approximately with F-2/3. Unexpectedly, over the energy range of 40 kVp to 18 MV, W +/- was found to decrease by a factor of nearly 3. These dependencies are compared to the predictions of two competing charge recombination models, geminate and columnar. The results are explained by a microdosimetric model in which the sensitivity at megavoltage energies is governed by geminate recombination, but at lower energies, both mechanisms are involved. Thus, the sensitivity of a-Se to x rays spanning the diagnostic and radiotherapy range has been measured and the physical basis for this behavior established.

Measurement of quantum noise in fluoroscopic systems for portal imaging.

In fluoroscopic portal imaging systems, a metal plate is bonded to a phosphor screen and together these act as the primary x-ray sensor. The light from the screen is collected and imaged by a lens on the target of a video camera. The demagnification (M) between the large area of the phosphor being imaged and the small active area of the video camera results in poor optical coupling between the screen and the video camera. Consequently x-ray quantum noise is small compared to other noise sources. By reducing the demagnification, the light from the screen is collected more efficiently, so we were able to increase the x-ray quantum noise relative to other noise sources and thus unambiguously identify it. The noise power spectrum was measured as a function of M to determine the relationship between the x-ray quantum noise. shot noise, and amplifier noise. It was found by extrapolation to clinical demagnifications that the amplifier noise dominates x-ray quantum noise, at all spatial frequencies, but the shot noise was less than the x-ray quantum noise at low spatial frequencies. For low spatial frequencies, this implies that a secondary quantum sink can be avoided. If amplifier noise could be sufficiently reduced, x-ray quantum limited images could be obtained in clinical systems at low spatial frequencies.

Low-energy imaging with high-energy bremsstrahlung beams: analysis and scatter reduction.

The contrast and zero spatial frequency signal-to-noise ratio produced by a method for radiation therapy portal imaging known as low-energy imaging with high-energy bremsstrahlung beams have been mathematically analyzed. The analysis makes extensive use of Monte Carlo techniques and incorporates the detector, the spectrum, phantom, and geometry. The analysis is validated through comparison with measured data including subject contrast measurements and the attenuation of the beam with lead. Scatter reduction is found to be potentially the most effective method to improve contrast and SNR for a film based system. A large fraction of the scatter detected is of a much higher energy than that found in diagnostic radiology. Hence, traditional antiscatter grids, such as those used in diagnostic radiology, are ineffective. The analysis and theory from the literature are applied to design a new grid which is more appropriate for this application. The grid produces a modest improvement according to a contrast-detail study.

Design of parallel plate ion chambers for buildup measurements in megavoltage photon beams.

Dose measurements in the buildup region of megavoltage photon beams are most commonly made using parallel plate ion chambers having fixed electrode separation. Fixed-separation chambers generally do not read correctly under such beam conditions because of the contribution to the chamber signal of electrons from the side walls. In this work it is shown that the side wall error can be very large and published correction formulas are not accurate for all beam conditions and chamber geometries. The principal focus of this study has been to determine the design features of a fixed-separation chamber that has negligible side wall error. The approach has been to study, in beams of 60Co, 6 MV, and 18 MV, the response of a specially built ion chamber in which several chamber parameters could be independently varied. The study has shown that the side wall error is primarily dependent on the ratio of the electrode separation to the wall diameter as well as on the wall density and wall angle. Based on these findings the design of a fixed-separation chamber is described which reads to within about 1% of the correct dose. Guidelines are also provided for assessing the suitability of current commercial fixed-separation ion chambers for buildup measurements.

Therapy imaging: a signal-to-noise analysis of a fluoroscopic imaging system for radiotherapy localization.

We have been developing a digital fluoroscopic imaging system to replace the portal films that are currently used to verify patient positioning during radiotherapy treatments. Our system differs from previously reported devices in the construction of the detector and in the operation of the TV camera. The signal, noise, and signal-to-noise properties of this system have been determined by measuring the modulation transfer function [MTF(f)], the noise power spectra [NPS(f)], and by calculating the detective quantum efficiency [DQE(f)] of the system. The results show: (i) that the spatial resolution of the system is determined largely by the lens of the TV camera and by frame grabber; and (ii) that the noise in the system is dominated by the secondary light quanta, due to the poor light collection efficiency of the optical chain. Despite these physical limitations, a contrast-detail study shows that the fluoroscopic system is better at detecting large, low contrast objects than portal films. Therefore the system is already a reasonable alternative to portal films and modifications to the metal plate/phosphor detector, lens, TV camera, and frame grabber should improve the performance of the system further.

A dose-per-pulse monitor for a dual-mode medical accelerator.

On a radiotherapy accelerator, the dose monitoring system is the last level of protection between the patient and the extremely high dose rate which all accelerators are capable of producing. The risk of losing this level of protection is substantially reduced if two or more dose monitoring systems are used which are mechanically and electrically independent in design. This paper describes the installation of an independent radiation monitor in a dual-mode, computer-controlled accelerator with a moveable monitor chamber. The added device is fixed in the beam path, is capable of monitoring each beam pulse, and is capable of terminating irradiation within the pulse repetition period if any measured pulse is unacceptably high.

A digital fluoroscopic imaging device for radiotherapy localization.

We have been developing a digital fluoroscopic imaging system to replace the portal films that are currently used to verify patient positioning during radiotherapy treatments. Our system has a number of modifications compared to previously reported devices. The detector, which consists of a copper plate with Gd2O2S:Tb phosphor bonded directly to the copper, has been designed to maximize light output from the phosphor by increasing the phosphor thickness. The operation of the T.V. camera has been modified so that the light signal is accumulated on the target of the T.V. camera for periods of 0.2-2.0 seconds. Accumulation of the light increases the video signal relative to the fixed noise current generated by the camera, and thus minimizes the camera noise. The resulting image quality is comparable to film, so the imaging system represents a promising alternative to film as a method of verifying patient positioning in radiotherapy.

Therapy imaging: source sizes of radiotherapy beams.

We have developed a novel method, which employs large lead collimators and computed tomography reconstruction techniques, to measure the intensity distributions of x-ray sources of radiotherapy devices. Using this method, we have measured the intensity distributions of x-ray sources from 60Co, 6-, 18-, and 25-MV radiotherapy devices. The x-ray sources of the accelerators were all elliptical in shape, but varied in eccentricity, and the sizes of the accelerator sources varied from 0.7 to 3.3 mm full width at half-maximum. The 60Co source was circular in shape and 20 mm in diameter, however, the output from this source was not uniform across its face. The modulation transfer functions (MTF's) (at the image plane) calculated for the accelerator sources, assuming an image magnification of 1.2, had similar magnitudes at low spatial frequencies as the MTF's of the metal plate/film detectors commonly used for therapy imaging. However, the source MTF's declined much more rapidly at high spatial frequencies. Therefore, for geometries commonly found in radiotherapy, the loss in spatial resolution due to the x-ray source was at least equal to that caused by electron and photon scatter within the metal plate/film detectors.

Therapy imaging: a signal-to-noise analysis of metal plate/film detectors.

We have measured the modulation transfer functions [MTF (f)'s] and the noise power spectra [NPS (f)] of therapy x-ray detectors irradiated by 60Co, 6- and 18-MV radiotherapy beams. Using these quantities, we have calculated the noise-equivalent quanta [NEQ (f)] and the detective quantum efficiency [DQE (f)] to quantitate the limitations of therapy detectors. The detectors consisted of film or fluorescent screen-film combinations in contact with copper, lead, or tungsten metal plates. The resolution of the detectors was found to be comparable to fluorescent screen-film combinations used in diagnostic radiology, however, the signal-to-noise ratio [SNR (f)] of the detectors was limited due to film granularity. We conclude that improved images can be obtained by using alternative detector systems which have less noise or film granularity.

A flatness and calibration monitor for accelerator photon and electron beams.

A flatness monitor has been built to quickly and accurately check accelerator beam flatness and dose calibration. Consisting of a 7 X 7 ion chamber array, the unit operates in photon beams from 60Co energies to 25 MV and electron beams (scattered or scanned) from 6 MeV to 25 MeV.

Direct measurement of electron contamination in cobalt beams using a charge detector.

A method is described in some detail for measuring the magnitude and penetration of the electron contamination in photon beams using a pancake charge detector. It is shown that the response of the detector to a photon beam can be separated from the component due to the electron contamination. In the present work, the detector is used to measure the electron fluence in a 60Co photon beam. This fluence is subsequently converted to dose by comparison with the fluence and dose measured from a pure electron beam (90Sr). This study proves, within experimental error, that the observed changes in the buildup region, with the collimator opening for both filtered and unfiltered 60Co beams, are due to electron, rather than photon, contamination.

Theoretical and experimental investigation of dose enhancement due to charge storage in electron-irradiated phantoms.

Recent measurements have shown that significant errors in radiation dosimetry can arise by the use of insulating plastic phantoms which have been exposed to electron beams. The effect has been attributed to the generation of large electric fields in the phantom by charge storage causing alteration of electron trajectories and an increase in the measured dose. In this report, we examine this hypothesis theoretically by calculating the change in response to radiation of an ion chamber in a cylindrical cavity in an electron-irradiated polymethylmethacrylate phantom. The electric field distribution is determined using a model which allows for charge leakage by radiation-induced conductivity, and the dose in the cavity is determined by a Monte Carlo simulation using the EGS (electron gamma shower) code modified to account for electron trajectories in the electric field. The theoretical results are shown to agree well with new and previously published experimental dose enhancement data. The agreement is taken as confirmation of the reported explanation of the effect. The use of conducting phantoms in radiation dosimetry is advocated.

Dose errors due to charge storage in electron irradiated plastic phantoms.

Commercial plastics used for radiation dosimetry are good electrical insulators . Used in electron beams, these insulators store charge and produce internal electric fields large enough to measurably alter the electron dose distribution in the plastic. The reading per monitor unit from a cylindrical ion chamber imbedded in a polymethylmethacrylate (PMMA) or polystyrene phantom will increase with accumulated electron dose, the increase being detectable after about 20 Gy of 6-MeV electrons. The magnitude of the effect also depends on the type of the plastic, the thickness of the plastic, the wall thickness of the detector, the diameter and depth of the hole in the plastic, the energy of the electron beam, and the dose rate used. Effects of charge buildup have been documented elsewhere for very low energy electrons at extremely high doses and dose rates. Here we draw attention to the charging effects in plastics at the dose levels encountered in therapy dosimetry where ion chamber or other dosimeter readings may easily increase by 5% to 10% and where a phantom, once charged, will also affect subsequent readings taken in 60Co beams and high-energy electron and x-ray beams for periods of several days to many months. It is recommended that conducting plastic phantoms replace PMMA and polystyrene phantoms in radiation dosimetry.

Partial bolussing to improve the depth doses in the surface region of low energy electron beams.

In many low energy electron beams the surface dose is considerably less than the maximum dose, making them unsatisfactory for clinical application. A method is described for producing better surface dose uniformity in such beams. The method makes use of bolus applied to the patient for a fraction of each daily electron treatment. The technique is shown to be simple and practical. This approach is compared to more conventional techniques of using bolus in electron beams.

Ontario accelerator dose intercomparison study.

An ionometric dose intercomparison has been carried out on eight accelertors, with maximum photon energies from 6 to 32 meV, at five radiotherapy centres in Ontario. The ratio of the dose based on the clinically-employed rad/monitor unit to the dose measured by the committee representative had a mean value of 0.994, with a range of 7.3% and a coefficient of variation of 2.5%. The ratio of the dose measured by the institution's physicist to the dose measured by the committee representative had a mean value of 1.000 with a range of 6.9% and a coefficient of variation of 2.3%. Eight recommendations regarding dose calibration procedures are presented.