Table of contents

Gamma cameras based on charge-coupled devices (CCDs) and micro-columnar CsI scintillators can reach high spatial resolutions. However, the gamma interaction probability of these scintillators is low (typically <30% at 141 keV) due to the limited thickness of presently available micro-columnar scintillators. Continuous scintillators can improve the interaction probability but suffer from increased light spread compared to columnar scintillators. In addition, for both types of scintillators, gamma photons incident at an oblique angle reduce the spatial resolution due to the variable depth of interaction (DOI). To improve the spatial resolution and spectral characteristics of these detectors, we have developed a fast analytic scintillation detection algorithm that makes use of a depth-dependent light spread model and as a result is able to estimate the DOI in the scintillator. This algorithm, performing multi-scale frame analysis, was tested for an electron multiplying CCD (EM-CCD) optically coupled to CsI(Tl) scintillators of different thicknesses. For the thickest scintillator (2.6 mm) a spatial resolution of 148 µm full width half maximum (FWHM) was obtained with an energy resolution of 46% FWHM for perpendicularly incident gamma photons (interaction probability 61% at 141 keV). The multi-scale algorithm improves the spatial resolution up to 11%, the energy resolution up to 36% and the signal-to-background counts ratio up to 46% compared to a previously implemented algorithm that did not model the depth-dependent light spread. In addition, the multi-scale algorithm can accurately estimate DOI. As a result, degradation of the spatial resolution due to the variable DOI for gamma photons incident at a 45° angle was improved from 2.0 ⋅ 10³ to 448 µm FWHM. We conclude that the multi-scale algorithm significantly improves CCD-based gamma cameras as can be applied in future SPECT systems.

Radiation-sensitive polymer gels for clinical dosimetry have been intensively investigated with magnetic resonance imaging (MRI) because the transversal magnetic relaxation time is dependent on irradiation dose. MRI is expensive and not easily available in most clinics. For this reason, low-cost, quick and easy-to-use potential alternatives such as optical computed tomography (CT), x-ray CT or ultrasound attenuation CT have also been studied by others. Here, we instead evaluate the dose dependence of the elastic material property, Young's modulus and the dose response of the viscous relaxation of radiation-sensitive gels to discuss their potential for dose imaging. Three batches of a radiation-sensitive polymer gel (MAGIC gel) samples were homogeneously irradiated to doses from 0 Gy to 45.5 Gy. Young's modulus was computed from the measured stress on the sample surface and the strain applied to the sample when compressing it axially, and the viscous relaxation was determined from the stress decay under sustained compression. The viscous relaxation was found not to change significantly with dose. However, Young's modulus was dose dependent; it approximately doubled in the gels between 0 Gy and 20 Gy. By fitting a second-order polynomial to the Young's modulus-versus-dose data, 99.4% of the variance in Young's modulus was shown to be associated with the change in dose. The precision of the gel production, irradiation and Young's modulus measurement combined was found to be 4% at 2 Gy and 3% at 20 Gy. Potential sources of measurement error, such as those associated with the boundary conditions in the compression measurement, inhomogeneous polymerization, temperature (up to 1% error) and the evaporation of water from the sample (up to 1% error), were estimated and discussed. It was concluded that Young's modulus could be used for dose determination. Imaging techniques such as elastography may help to achieve this if they can provide a local measurement of Young's modulus, which may eliminate problems associated with the boundaries (e.g. variation in coefficient of friction) and inhomogeneous polymerization. Elastography combined with a calibration should also be capable of mapping dose in three dimensions.

The imaging performance of phosphor screens, used as x-ray detectors in diagnostic medical imaging systems, is affected by their both noise and resolution properties. Amplification and blurring processes are due to a sequence of conversion stages within the screen which contribute to fluctuations in the number and spatial distribution of the optical quanta recorded by the optical detector (e.g. film, television camera, CCD, etc). The purpose of this paper is to investigate the stochastic noise arising from granularity as well as the variation of spatial resolution of granular fluorescent screens in terms of the detector's structure. Using a custom-validated Monte Carlo model, the parameters of interest were evaluated for the widely used Gd₂O₂S:Tb phosphor material. We have studied the variations of (i) the modulation transfer function, (ii) the Swank factor and (iii) the zero-frequency detective quantum efficiency (DQE), under several conditions employed in conventional and digital mammography and radiology. Several evaluations are provided for the imaging metrics as a function of the x-ray energy (18 keV, 49 keV and 51 keV), phosphor coating weight (20 mg cm⁻², 34 mg cm⁻² and 60 mg cm⁻²), grain size (from 4 µm up to 13 µm) and packing density (from 50% up to 85%). It was found that screens of high packing density can combine high zero-frequency DQE with improved resolution properties. For a digital mammographic imaging system (34 mg cm⁻², 18 keV), a packing density of 85% can improve the spatial resolution of the screen by 1.6 cycles mm⁻¹ in comparison to that of 50% packing density. Similarly, for radiographic cases (60 mg cm⁻², 49 keV), the spatial resolution can be improved by 1.7 cycles mm⁻¹. The aforementioned findings provide the resolution benefits of using high packing density screens.

The reference levels for testing compliance of human exposure with radio-frequency (RF) safety limits have been derived from very simplified models of the human. In order to validate these findings for anatomical models, we investigated the absorption characteristics for various anatomies ranging from 6 year old child to large adult male by numerical modeling. We address the exposure to plane-waves incident from all major six sides of the humans with two orthogonal polarizations each. Worst-case scattered field exposure scenarios have been constructed in order to test the implemented procedures of current in situ compliance measurement standards (spatial averaging versus peak search). Our findings suggest that the reference levels of current electromagnetic (EM) safety guidelines for demonstrating compliance as well as some of the current measurement standards are not consistent with the basic restrictions and need to be revised.

Small animal research allows detailed study of biological processes, disease progression and response to therapy with the potential to provide a natural bridge to the clinical environment. The small animal radiation research platform (SARRP) is a portable system for precision irradiation with beam sizes down to approximately 0.5 mm and optimally planned radiation with on-board cone-beam CT (CBCT) guidance. This paper focuses on the geometric calibration of the system for high-precision irradiation. A novel technique for the calibration of the treatment beam is presented, which employs an x-ray camera whose precise positioning need not be known. Using the camera system we acquired a digitally reconstructed 3D 'star shot' for gantry calibration and then developed a technique to align each beam to a common isocenter with the robotic animal positioning stages. The calibration incorporates localization by cone-beam CT guidance. Uncorrected offsets of the beams with respect to the calibration origin ranged from 0.4 mm to 5.2 mm. With corrections, these alignment errors can be reduced to the sub-millimeter range. The calibration technique was used to deliver a stereotactic-like arc treatment to a phantom constructed with EBT Gafchromic films. All beams were shown to intersect at a common isocenter with a measured beam (FWHM) of approximately 1.07 mm using the 0.5 mm collimated beam. The desired positioning accuracy of the SARRP is 0.25 mm and the results indicate an accuracy of 0.2 mm. To fully realize the radiation localization capabilities of the SARRP, precise geometric calibration is required, as with any such system. The x-ray camera-based technique presented here provides a straightforward and semi-automatic method for system calibration.

A factor currently limiting the clinical utility of x-ray CT polymer gel dosimetry is the overall low dose sensitivity (and hence low dose resolution) of the system. Hence, active research remains in the investigation of polymer gel formulations with increased CT dose response. An ideal polymer gel dosimeter will exhibit a sensitive CT response which is linear over a suitable dose range, making clinical implementation reasonably straightforward. This study reports on the variations in rate and form of the CT dose response of irradiated polymer gels manufactured with glycerol, which is a co-solvent that permits dissolution of additional bisacrylamide above its water solubility limit (3% by weight). This study focuses on situations where the concentration of bisacrylamide is kept at or below its water solubility limit so that the influence of the co-solvent on the dose response can be explored separately from the effects of increased cross-linker concentration. CT imaging and Raman spectroscopy are used to construct dose–response curves for irradiated gels varying in (i) initial total monomer (%T) and (ii) initial co-solvent concentration. Results indicate that: (i) for a fixed glycerol concentration, gel response increases linearly with %T. Furthermore, the functional form of the dose response remains constant, in agreement with a previous model of polymer formation. (ii) Polymer gels with constant %T and increasing co-solvent concentrations also show enhanced CT response. In addition, the functional form of the response is altered in these gels as co-solvent concentration is increased. Raman data indicate that the fraction of bis-acrylamide incorporated into polymerization, as opposed to cyclization, increases as co-solvent concentration increases. The changes in functional form indicate varying polymer yields (per unit dose), akin to relative fractional monomer/cross-linker (i.e. %C) changes in earlier studies. These results are put into context of the model of polymer formation. The implications of these results on the clinical utility of polymer gels with co-solvent are highlighted.

The use of least-squares regression to probe the level of lead contamination of plaster of Paris standards in the calibration of ¹⁰⁹Cd KXRF systems for bone lead measurement, as well as the use of iteratively reweighted least-squares (IRLS) in the case of violation of the assumptions for ordinary least-squares (OLS), is discussed here. One common violation is non-uniform residual variance, which makes the use of OLS inappropriate due to strong influence of points with large variance on the calibration line and variance of the slope and intercept. Comparison between OLS and IRLS in that case showed that IRLS estimates of the intercept are significantly smaller and more precise than OLS estimates, while a less marked increase in the calibration slope is observed when IRLS is used. Moreover, OLS underestimates bone lead concentrations at low levels of lead exposure and overestimates those concentrations at higher levels. These discrepancies are smaller in magnitude than the measurement uncertainty of conventional systems, except for high concentrations. For the newly developed cloverleaf systems, the suggested differences at bone lead concentrations below 17 ppm are comparable to the minimum detection limit, but are larger than the measurement uncertainty for bone lead concentrations above 60 ppm.

We present a method to incorporate geometrical uncertainties into dose–volume histogram evaluation: the dose–volume population histogram (DVPH). For each dose–volume point, the probability of the plan DVH meeting the constraint is calculated. The use of DVPH for the target shows that the minimum dose to the PTV might not be representative of the minimum dose to the CTV considering geometrical uncertainties when the PTV extends into the build-up region. For OARs, the DVH obtained from DVPH with 90% confidence level is found to be different to the planning organ at risk volume (PRV) DVH recommended by ICRU 62, especially for parallel organs.

This study aims to assess three methods commonly used in content-based image retrieval (CBIR) schemes and investigate the approaches to improve scheme performance. A reference database involving 3000 regions of interest (ROIs) was established. Among them, 400 ROIs were randomly selected to form a testing dataset. Three methods, namely mutual information, Pearson's correlation and a multi-feature-based k-nearest neighbor (KNN) algorithm, were applied to search for the 15 'the most similar' reference ROIs to each testing ROI. The clinical relevance and visual similarity of searching results were evaluated using the areas under receiver operating characteristic (ROC) curves (A_Z) and average mean square difference (MSD) of the mass boundary spiculation level ratings between testing and selected ROIs, respectively. The results showed that the A_Z values were 0.893 ± 0.009, 0.606 ± 0.021 and 0.699 ± 0.026 for the use of KNN, mutual information and Pearson's correlation, respectively. The A_Z values increased to 0.724 ± 0.017 and 0.787 ± 0.016 for mutual information and Pearson's correlation when using ROIs with the size adaptively adjusted based on actual mass size. The corresponding MSD values were 2.107 ± 0.718, 2.301 ± 0.733 and 2.298 ± 0.743. The study demonstrates that due to the diversity of medical images, CBIR schemes using multiple image features and mass size-based ROIs can achieve significantly improved performance.

A Monte Carlo simulation of repeated cubic units representing trabecular bone cavities in adult bone was employed to determine absorbed dose fractions evaluated for ³H, ¹⁴C and a set of α-emitters incorporated within a bone remodeling compartment (BRC). The BRC consists of a well-oxygenated vascular microenvironment located within a canopy of bone-lining cells. The International Commission on Radiological Protection (ICRP) considers that an important target for radiation-induced bone cancer is the endosteum marrow layer adjacent to bone surface where quiescent bone stem cells reside. It is proposed that the active stem cells and progenitor cells located above the BRC canopy, the 'BRC stem cell niche', is a more important radiation-induced cancer target volume. Simulation results from a static model, where no remodeling occurs, indicate that the mean dose from bone and bone surface to the 50 µm quiescent bone stem cell niche, the current ICRP target, was substantially lower (two to three times lower) than that to the narrower and hypoxic 10 µm endosteum for ³H, ¹⁴C and α-particles with energy range 0.5–10 MeV. The results from a dynamic model indicate that the temporal α-radiation dose to active stem/progenitor cells located in the BRC stem cell niche from the material incorporated in and buried by forming bone was 9- to 111-fold greater than the dose to the quiescent bone stem cell niche. This work indicates that the remodeling portion of the bone surface, rather than the quiescent (endosteal) surface, has the greatest risk of radiation-induced bone cancer, particularly from short-range radiation, due to the elevated dose and the radiosensitizing oxygen effect.

Accurate lung tumor tracking in real time is a keystone to image-guided radiotherapy of lung cancers. Existing lung tumor tracking approaches can be roughly grouped into three categories: (1) deriving tumor position from external surrogates; (2) tracking implanted fiducial markers fluoroscopically or electromagnetically; (3) fluoroscopically tracking lung tumor without implanted fiducial markers. The first approach suffers from insufficient accuracy, while the second may not be widely accepted due to the risk of pneumothorax. Previous studies in fluoroscopic markerless tracking are mainly based on template matching methods, which may fail when the tumor boundary is unclear in fluoroscopic images. In this paper we propose a novel markerless tumor tracking algorithm, which employs the correlation between the tumor position and surrogate anatomic features in the image. The positions of the surrogate features are not directly tracked; instead, we use principal component analysis of regions of interest containing them to obtain parametric representations of their motion patterns. Then, the tumor position can be predicted from the parametric representations of surrogates through regression. Four regression methods were tested in this study: linear and two-degree polynomial regression, artificial neural network (ANN) and support vector machine (SVM). The experimental results based on fluoroscopic sequences of ten lung cancer patients demonstrate a mean tracking error of 2.1 pixels and a maximum error at a 95% confidence level of 4.6 pixels (pixel size is about 0.5 mm) for the proposed tracking algorithm.

In this work the neutron production in a passive beam delivery system was investigated. Secondary particles including neutrons are created as the proton beam interacts with beam shaping devices in the treatment head. Stray neutron exposure to the whole body may increase the risk that the patient develops a radiogenic cancer years or decades after radiotherapy. We simulated a passive proton beam delivery system with double scattering technology to determine the neutron production and energy distribution at 200 MeV proton energy. Specifically, we studied the neutron absorbed dose per therapeutic absorbed dose, the neutron absorbed dose per source particle and the neutron energy spectrum at various locations around the nozzle. We also investigated the neutron production along the nozzle's central axis. The absorbed doses and neutron spectra were simulated with the MCNPX Monte Carlo code. The simulations revealed that the range modulation wheel (RMW) is the most intense neutron source of any of the beam spreading devices within the nozzle. This finding suggests that it may be helpful to refine the design of the RMW assembly, e.g., by adding local shielding, to suppress neutron-induced damage to components in the nozzle and to reduce the shielding thickness of the treatment vault. The simulations also revealed that the neutron dose to the patient is predominated by neutrons produced in the field defining collimator assembly, located just upstream of the patient.

Diffusion tensor tractography (DTT) allows one to explore axonal connectivity patterns in neuronal tissue by linking local predominant diffusion directions determined by diffusion tensor imaging (DTI). The majority of existing tractography approaches use continuous coordinates for calculating single trajectories through the diffusion tensor field. The tractography algorithm we propose is characterized by (1) a trajectory propagation rule that uses voxel centres as vertices and (2) orientation probabilities for the calculated steps in a trajectory that are obtained from the diffusion tensors of either two or three voxels. These voxels include the last voxel of each previous step and one or two candidate successor voxels. The precision and the accuracy of the suggested method are explored with synthetic data. Results clearly favour probabilities based on two consecutive successor voxels. Evidence is also provided that in any voxel-centre-based tractography approach, there is a need for a probability correction that takes into account the geometry of the acquisition grid. Finally, we provide examples in which the proposed fibre-tracking method is applied to the human optical radiation, the cortico-spinal tracts and to connections between Broca's and Wernicke's area to demonstrate the performance of the proposed method on measured data.

The application of a photoacoustic imaging instrument based upon a Fabry–Perot polymer film ultrasound sensor to imaging the superficial vasculature is described. This approach provides a backward mode-sensing configuration that has the potential to overcome the limitations of current piezoelectric based detection systems used in superficial photoacoustic imaging. The system has been evaluated by obtaining non-invasive images of the vasculature in human and mouse skin as well as mouse models of human colorectal tumours. These studies showed that the system can provide high-resolution 3D images of vascular structures to depths of up to 5 mm. It is considered that this type of instrument may find a role in the clinical assessment of conditions characterized by changes in the vasculature such as skin tumours and superficial soft tissue damage due to burns, wounds or ulceration. It may also find application in the characterization of small animal cancer models where it is important to follow the tumour vasculature over time in order to study its development and/or response to therapy.

A compression paddle is always used in mammography x-ray examinations, in order to improve image quality and reduce patient doses. Although clinical dose measurements should be performed with the paddle to interfere with the x-ray beam, calibration of mammography dosimeters is performed free in air without the presence of the paddle. The paddle hardens the x-ray beam, which has an impact on a dosimeter performance, particularly on high-energy-dependent detectors. Due to the paddle, clinical mammography x-ray systems may exhibit beams with HVL values exceeding those of the IEC 61267 RQR-M series qualities at which dosimeters are usually calibrated. In this study, the influence of the paddle in mammography dosimetry is examined, in Mo/Mo anode/filter x-ray qualities. PMMA slabs of 1, 2 and 3 mm thickness and Al foils of 0.05, 0.10 and 0.15 mm thicknesses were used to simulate the paddles, producing beams with HVL values from 0.28 up to 0.43 mmAl. In these qualities, four solid-state (ST) detectors and three ionizations chambers (IC) were calibrated in terms of Kair and N_K and k_Q were deduced. The results showed that all IC and two modern-type ST dosimeters have a flat energy response in the above HVL range (less than 3%), so their calibration factor at RQR-M2 quality could be safely used for clinical measurements. Two other ST dosimeters exhibit up to 20% energy response, so differences up to 15% in dose measurement may be observed if the effect of paddle on their performance is ignored. Finally, the need of additional mammographic calibration qualities to the existing IEC 61267 RQR-M series is examined and discussed.

We present a faster iterative reconstruction algorithm based on the ordered-subset convex (OSC) algorithm for transmission CT. The OSC algorithm was modified such that it calculates the normalization term before the iterative process in order to save computational cost. The modified version requires only one backprojection per iteration as compared to two required for the original OSC. We applied the modified OSC (MOSC) algorithm to a rotation-free micro-CT system that we proposed previously, observed its performance, and compared with the OSC algorithm for 3D cone-beam reconstruction. Measurements on the reconstructed images as well as the point spread functions show that MOSC is quite similar to OSC; in noise-resolution trade-off, MOSC is comparable with OSC in a regular-noise situation and it is slightly worse than OSC in an extremely high-noise situation. The timing record shows that MOSC saves 25–30% CPU time, depending on the number of iterations used. We conclude that the MOSC algorithm is more efficient than OSC and provides comparable images.

Gafchromic EBT (EBT) films are becoming increasingly popular due to their advantageous properties. When flatbed colour scanners are used for film dosimetry, a good quality control of the scanning device is a crucial step for accurate results. The proposal of this work was to fully assess the performance of the scanner Epson Expression 10000XL in order to quantify all parameters and needed corrections to minimize dose uncertainties. A standard step tablet, with 32 steps and optical densities from 0.06 to 3.8, was used to check the scanner linearity. The scanner warming-up effect and reproducibility were evaluated by performing 30 consecutive scans plus 20 scans in 15 min intervals. The scanning colour modes: 24 and 48 bits and scanning resolutions from 50 to 300 dpi were tested. A Wiener filter with different pixels regions was applied with the purpose of reducing the film noise. All scans were made in transmission mode with a constant film orientation. The red colour channel was posteriorly extracted from the images to maximize readout sensitivity. Two EBT films were irradiated, perpendicularly and parallel to beam incidence, with a 6 MV photon beam with doses that ranged from 0.2 to 3 Gy. A polynomial expression was used to convert optical density into dose. Dose uncertainty was quantified applying error propagation analysis. A correction for the non-uniform response of the scanner was determined using five films irradiated with a uniform dose. The scanner response was linear until an optical density of approximately 1 which corresponds to doses higher than those of clinical interest for EBT films. The scanner signal stabilized after seven readings. Scanner reproducibility around 0.2% was obtained either with the scanner warm or cold. However, reproducibility was significantly reduced when comparing images digitized with the scanner at different temperatures. Neither the colour depth mode, the scanning resolution, the multiscan option nor the Wiener filter had a significant effect on the shape of the calibration curve. However, a reduction in dose uncertainty was possible by selecting appropriate reading parameters. These are a 48 bit colour depth, a 75 dpi resolution and repeating the scan four times. Finally, the two dimensional Wiener filter applied to a 3 × 3 pixel region to the red component of the image reduced the experimental scan uncertainty to about 0.5% for doses higher than 0.5 Gy. Total scan uncertainty was less than 2% for a perpendicular calibration and reduced to less than 1% for a parallel calibration. A dose over-estimation of around 5% for clinical doses may be made if the image acquired is not corrected for the non-uniform response of the scanner. A protocol to read EBT films using the Epson Expression 10000XL scanner was established for IMRT verification. The contribution for the overall uncertainty in film dosimetry coming from the scanning process was estimated to be around 0.5% for doses higher than 0.5 Gy when reading parameters are optimized. Total scan uncertainty achieved is about 2% when using a perpendicular calibration. It can further be reduced if a parallel calibration is used.

There is a serious and growing concern about the increased risk of radiation-induced second cancers and late tissue injuries associated with radiation treatment. To better understand and to more accurately quantify non-target organ doses due to scatter and leakage radiation from medical accelerators, a detailed Monte Carlo model of the medical linear accelerator is needed. This paper describes the development and validation of a detailed accelerator model of the Varian Clinac operating at 6 and 18 MV beam energies. Over 100 accelerator components have been defined and integrated using the Monte Carlo code MCNPX. A series of in-field and out-of-field dose validation studies were performed. In-field dose distributions calculated using the accelerator models were tuned to match measurement data that are considered the de facto 'gold standard' for the Varian Clinac accelerator provided by the manufacturer. Field sizes of 4 cm × 4 cm, 10 cm × 10 cm, 20 cm × 20 cm and 40 cm × 40 cm were considered. The local difference between calculated and measured dose on the percent depth dose curve was less than 2% for all locations. The local difference between calculated and measured dose on the dose profile curve was less than 2% in the plateau region and less than 2 mm in the penumbra region for all locations. Out-of-field dose profiles were calculated and compared to measurement data for both beam energies for field sizes of 4 cm × 4 cm, 10 cm × 10 cm and 20 cm × 20 cm. For all field sizes considered in this study, the average local difference between calculated and measured dose for the 6 and 18 MV beams was 14 and 16%, respectively. In addition, a method for determining neutron contamination in the 18 MV operating model was validated by comparing calculated in-air neutron fluence with reported calculations and measurements. The average difference between calculated and measured neutron fluence was 20%. As one of the most detailed accelerator models for both in-field and out-of-field dose calculations, the model will be combined with anatomically realistic computational patient phantoms into a computational framework to calculate non-target organ doses to patients from various radiation treatment plans.

The present study investigated the whole-body averaged specific absorption rate (WBSAR) in an infant model with the finite-difference time-domain method. The focus of the present study is the effect of polarization of incident electromagnetic waves on the WBSAR. This is because most previous studies investigated the WBSAR for plane-wave exposure with a vertically aligned electric field. Our computational results revealed that the WBSAR for plane-wave exposure with a vertically aligned electric field is smaller than that with a horizontally aligned electric field for frequencies above 2 GHz. The main reason for this difference is attributed to be the component of the surface area perpendicular to the electric field of the incident wave.

Parts (a) and (b) of figure 8 in this article were in error not included in the published article. The correct figure is reproduced below.

Figure 8. Plots of CRC (a) and (b) and Noise (c) and (d) as a function of iteration number for varying scan times in the 35 cm lesion detectability phantom. Left column (a and c) shows result from TOF reconstructions while the right column (b) and (d) has the non-TOF reconstruction results. The curves within each plot are for varying scan times of 2 (○), 3 (▿), 4 ◊, and 5 minutes □.