A. Calzado

Validation of a Monte Carlo simulation for dose assessment in dental cone beam CT examinations

A Monte Carlo (MC) simulation for calculating absorbed dose has been developed and applied for dental applications with an i-CAT cone beam CT (CBCT) system. To validate the method a comparison was made between calculated and measured dose values for two different clinical protocols. Measurements with a pencil CT chamber were performed free-in-air and in a CT dose head phantom; measurements were also performed with a transmission ionization chamber. In addition for each protocol a total number of 58 thermoluminescence dosemeters (TLD) were packed in groups and placed at 16 representative anatomical locations of an anthropomorphic phantom (Remab system) to assess absorbed doses. To simulate X-ray exposure, a software application based on the EGS4 package was applied. Dose quantities were calculated for different voxelized models representing the CT ionization and transmission chambers, the TLDs, and the phantoms as well. The dose quantities evaluated in the comparison were the accumulated dose averaged along the rotation axis (D(i)), the volume average dose,D(vol) for the dosimetric phantom, the dose area product (DAP) and the absorbed dose for the TLDs. Absolute differences between measured and simulated outcomes were ≤ 2.1% for free-in-air doses; ≤ 6.2% in the 5 cavities of the CT dose head phantom; ≤ 13% for TLDs inside the primary beam. Such differences were considered acceptable in all cases and confirmed the validity of the MC program for different geometries. In conclusion, the devised MC simulation program can be a robust tool to optimize protocols and estimate patient doses for CBCT units in dental, oral and maxillofacial radiology.

Comparison between human and model observer performance in low-contrast detection tasks in CT images: application to images reconstructed with filtered back projection and iterative algorithms

OBJECTIVE To compare low-contrast detectability (LCDet) performance between a model [non-pre-whitening matched filter with an eye filter (NPWE)] and human observers in CT images reconstructed with filtered back projection (FBP) and iterative [adaptive iterative dose reduction three-dimensional (AIDR 3D; Toshiba Medical Systems, Zoetermeer, Netherlands)] algorithms. METHODS Images of the Catphan® phantom (Phantom Laboratories, New York, NY) were acquired with Aquilion ONE™ 320-detector row CT (Toshiba Medical Systems, Tokyo, Japan) at five tube current levels (20-500 mA range) and reconstructed with FBP and AIDR 3D. Samples containing either low-contrast objects (diameters, 2-15 mm) or background were extracted and analysed by the NPWE model and four human observers in a two-alternative forced choice detection task study. Proportion correct (PC) values were obtained for each analysed object and used to compare human and model observer performances. An efficiency factor (η) was calculated to normalize NPWE to human results. RESULTS Human and NPWE model PC values (normalized by the efficiency, η = 0.44) were highly correlated for the whole dose range. The Pearsons product-moment correlation coefficients (95% confidence interval) between human and NPWE were 0.984 (0.972-0.991) for AIDR 3D and 0.984 (0.971-0.991) for FBP, respectively. Bland-Altman plots based on PC results showed excellent agreement between human and NPWE [mean absolute difference 0.5 ± 0.4%; range of differences (-4.7%, 5.6%)]. CONCLUSION The NPWE model observer can predict human performance in LCDet tasks in phantom CT images reconstructed with FBP and AIDR 3D algorithms at different dose levels. ADVANCES IN KNOWLEDGE Quantitative assessment of LCDet in CT can accurately be performed using software based on a model observer.

Tomografía computarizada multicorte en un servicio de radiodiagnóstico: estudio de las dosis impartidas durante 1 año

OBJECTIVE To determine the usage of CT in our diagnostic imaging department and the dose of radiation CT delivered to patients. MATERIAL AND METHODS We carried out two studies, a retrospective study on the administrative data from one year and a prospective study over one month, to determine the amount of activity of the two multislice CT scanners in our department. We recorded the total number of examinations, the different types of examination performed, and the age distributions of the patients examined. We used these data and the results of dosimetric measurements to estimate the typical radiation dose for each type of examination, its contribution to the collective radiation dose, and the total collective radiation dose from CT examinations at our hospital. RESULTS The retrospective study found that 15038 examinations were performed in 12678 patients in one year. The most common CT examinations were cranial (38.2%), abdomen-pelvis (17.5%), chest-abdomen (11.7%), chest (9.3%), and lumbar-sacral spine (6.8%). Some protocols used doses of radiation that were higher than those recommended by European experts. The total collective dose of radiation delivered to patients at our hospital by CT was estimated at 70 person.Sv. Abdomen-pelvis examinations contributed 31% of the total dose. CONCLUSIONS This study has provided detailed knowledge about the use of CT in our diagnostic imaging department, about the dose of radiation delivered, and about possibilities for improving the use of our CT scanners.

Low contrast detectability performance of model observers based on CT phantom images: kVp influence.

This paper studies low contrast detectability (LCD) performance of two model observers in CT phantom images acquired at different kVp levels and compares the results with humans in a 2-alternative forced choice experiment (2-AFC). Images of the Catphan phantom with objects of different contrasts (0.5 and 1%) and diameters (2-15 mm) were acquired in an Aquilion ONE 320-detector row CT (Toshiba Medical Systems, Tokyo, Japan), in two experiments, selecting (80-100-120-135 kV) with fixed mAs and varying the mAs to keep the dose constant, respectively. Four human observers evaluated the objects visibility obtaining a proportion correct (PC) for each case. LCD was also analyzed with two model observers (non-prewhitening matched filter with an eye filter, NPWE, and channelized Hotelling observer with Gabor channels, CHO). Object contrast was affected by kV, with differences up to 17% between the lowest and highest kV. Both models overestimated human performance and were corrected by efficiency and internal noise factors. The NPWE model reproduced better the human PC values trends showing Pearsons correlation coefficients ≥0.976 (0.954-0.987, 95% CI) for both experiments, whereas for CHO they were ≥0.706 (0.493-0.839). Bland-Altman plots showed better agreement between NPWE and humans being the average difference Δ and the range of the differences Δ±2σ (σ, standard deviation) of Δ=-0.3%, Δ±2σ = [-4.0%,4.5%]. For CHO, Δ=-1.2%, Δ± 2σ= [-10.7%,8.3%]. The NPWE model can be a useful tool to predict human performance in CT low contrast detection tasks in a standard phantom and be potentially used in protocol optimization based on kV selection.

A parametrization of the CT number of a substance and its use for stoichiometric calibration

A method to carry out stoichiometric calibrations of CT scanners employed in radiotherapy treatment planning is proposed. The method is based on a simple parametrization of the CT number of a substance, which involves only two variables to describe the substance (electron density and one effective atomic number) and one parameter to describe the beam. The method was tested experimentally on a group of beams. A set of no tissue-like substances of known densities and elemental compositions were employed as calibrators. CT number-to-density curves (RED curves) were calculated with the proposed parametrization and compared to those measured with a commercial density phantom. Differences between the electron densities assigned by the calculated RED curves and the measured ones were in the range 0.009-0.019 (RMS). The proposed method may be employed to carry out accurate stoichiometric calibrations by using only one suitable substance as calibrator, not necessarily tissue-like.

A Monte Carlo simulation for the estimation of patient dose in rest and stress cardiac computed tomography with a 320-detector row CT scanner

PURPOSE To estimate organ dose and effective dose for patients for cardiac CT as applied in an international multicenter study (CORE320) with a 320-Detector row CT scanner using Monte Carlo (MC) simulations and voxelized phantoms. The effect of positioning of the arms, off-centering the patient and heart rate on patient dose was analyzed. METHODS A MC code was tailored to simulate the geometry and characteristics of the CT scanner. The phantoms representing the adult reference male and female were implemented according to ICRP 110. Effective dose and organ doses were obtained for CT acquisition protocols for calcium scoring, coronary angiography and myocardial perfusion. RESULTS For low heart rate, the normalized effective dose (E) for cardiac CT was higher for female (5.6 mSv/100 mAs) compared to male (2.2 mSv/100 mAs) due to the contribution of female breast tissue. Averaged E for female and male was 11.3 mSv for the comprehensive cardiac protocol consisting of calcium scoring (1.9 mSv); coronary angiography including rest cardiac perfusion (5.1 mSv) and stress cardiac perfusion (4.3 mSv). These values almost doubled at higher heart rates (20.1 mSv). Excluding the arms increased effective dose by 6-8%, centering the patient showed no significant effect. The k-factor (0.028 mSv/mGy.cm) derived from this study leads to effective doses up to 2-3 times higher than the values obtained using now outdated methodologies. CONCLUSION MC modeling of cardiac CT examinations on realistic voxelized phantoms allowed us to assess patient doses accurately and we derived k-factors that are well above those published previously.

42 MeV bremsstrahlung spectrum analysis by a photoactivation method

Abstract The evaluation of 42 MeV bremsstrahlung spectra from a clinical betatron by using the photoactivation method is described. Photonuclear reactions, mainly of the (γ, n) type, are used as activation detectors. After measurements of photon-induced activities from residual nuclei are performed, the spectral distribution of photons is evaluated by solving the unfolding problem. The latter is carried out through the use of two independent methods, orthonormal expansion and Monte Carlo. In both cases prior conditions to the solution are imposed. Spectra evaluated by both methods and making use of two different size flattening filters are presented. An empirical method to estimate the “effective” thickness of the Pt target is described.

Role of computed tomography on large B-cell non-Hodgkin’s lymphoma follow-up and the risk of radiation-induced neoplasm: A retrospective cohort study

INTRODUCTION To evaluate the role of surveillance computed tomography (CT) in detection of tumor relapse after complete remission for large-cell lymphoma and to estimate the cancer risk related to those examinations. MATERIALS AND METHODS We retrospectively identified the cases of large-cell lymphoma diagnosed at our institution during a fifteen years period. We extracted from charts the information needed. Agreement between clinical and radiological evaluation for relapse detection was determined with index kappa and McNemar tests. We analysed the surveillance CT examinations performed to the patients in complete remission with no recurrence during the follow-up period and we estimated the life attributable risk (LAR) of radiation-induced cancer above base line using the approach of the Biological Effects of Ionizing Radiation (BEIR) VII. RESULTS 184 patients with biopsy confirmed large-cell lymphoma were included. Complete remission was attained in 125 patients. After a median follow-up of 93.73 months, 97 of them remain in remission. Relapse was confirmed in 28 patients; only 3 of them were asymptomatic. Kappa and McNemar analyses revealed good agreement for diagnosis of recurrence and significant difference for exclusion of relapse. Patients in remission received a median of 6 surveillance CT examinations. Fifty-seven patients had total cumulative doses greater than 100mSv. The mean increase in estimated radiation-induced cancer risk was 0.5%. CONCLUSION Our results suggest that periodic CT examinations have a limited role in detecting relapse in large-cell lymphoma as the clinical surveillance dictates the patient management.

Validation of a MC code to assess patient doses from cone beam CT in dentistry

A Monte Carlo (MC) -based method to calculate absorbed dose from cone beam CT examinations has been developed and applied. To simulate X-ray exposure, a code using the EGS4 package and different voxel models represent- ing an ionization chamber, a dosimetry head phantom and an anthropomorphic phantom (Remab system) were employed. In order to validate the method, dose measurements free-in-air, and within the phantoms were performed using an ionization chamber and 58 thermoluminescence dosemeter (TLD) chips. TLD dosemeters were grouped in stacks of 5, 3 or 2 elements at 16 positions in the anthropomorphic phantom. A compari- son of the dose outcomes from both methods gave relative differences below 9% for the dose free-in-air, below 4.5% for the dose in the 9 cavities of the head phantom. For TLD meas- urements, relative differences in dose between both methods were i 15% for dosemeters in the primary beam. Differences in other TLD positions can be explained by critical dependence of the calculated dose on the geometrical accuracy. The MC code gives an acceptable description of the dose distribution for simple and anthropomorphic geometries in combination with voxel models of the object or the patient. From the simulation of the standard examination on the Remab phantom, values of average dose per slab (ADS), ener- gy imparted and mean absorbed organ doses were calculated. ADS values in the pituitary area were about 2 mGy, well cor- related with energy imparted per slab. Calculated mean ab- sorbed doses for selected organs were in the range 0.076 mGy (thyroid) to 1.74 mGy (salivary glands). In conclusion, the devised MC method has been validated and can be a robust tool to optimize protocols and estimate patient doses for a CBCT unit in maxillofacial radiology.