Object and person tracking systems to enable extremity dosimetry in nuclear medicine using computational methods.

Nuclear medicine (NM) professionals are potentially exposed to high doses of ionising radiation, particularly in the skin of the hands. Ring dosimeters are used by the workers to ensure extremity doses are kept below the legal limits. However, ring dosimeters are often susceptible to large uncertainties, so it is difficult to ensure a correct measurement using the traditional occupational monitoring methods. An alternative solution is to calculate the absorbed dose by using Monte Carlo simulations. This method could reduce the uncertainty in dose calculation if the exact positions of the worker and the radiation source are represented in these simulations. In this study we present a set of computer vision and artificial intelligence algorithms that allow us to track the exact position of unshielded syringes and the hands of NM workers. We showcase a possible hardware configuration to acquire the necessary input data for the algorithms. And finally, we assess the tracking confidence of our software. The tracking accuracy achieved for the syringe detection was 57% and for the hand detection 98%.

Effectiveness of staff radiation protection devices for interventional cardiology procedures.

PURPOSE: To evaluate the effectiveness of currently available radioprotective (RP) devices in reducing the dose to interventional cardiology staff, especially to the eye lens and brain. METHODS: The performances of five RP devices (masks, caps, patient drapes, staff lead and lead-free aprons and Zero-Gravity (ZG) suspended radiation protection system) were assessed by means of Monte Carlo (MC) simulations. A geometry representative of an interventional cardiology setup was modelled and several configurations, including beam projections and staff distance from the source, were investigated. In addition, measurements on phantoms were performed for masks and drapes. RESULTS: An average dose reduction of 65% and 25% to the eyes and the brain respectively was obtained for the masks by MC simulations but a strong influence of the design was observed. The cap effectiveness for the brain ranges on average between 13% and 37%. Nevertheless, it was shown that only some upper parts of the brain were protected. There was no significant difference between the effectiveness of lead and lead-free aprons. Of all the devices, the ZG system offered the highest protection to the brain and eye lens and a protection level comparable to the apron for the organs normally covered. CONCLUSION: All investigated devices showed potential for dose reduction to specific organs. However, for masks, caps and drapes, it strongly depends on the design, exposure conditions and staff position. Therefore, for a clinical use, it is recommended to evaluate their effectiveness in the planned conditions of use.

Impact of the implementation of the new radiation quantities recommended by ICRU/ICRP for practical use in interventional radiology: a Monte Carlo study.

The International Commission on Radiation Units and Measurements (ICRU) proposed a new set of operational quantities for radiation protection for external radiation in its Report Committee 26 (ICRU95). The new proposal aims to improve the coherence between the operational quantities and the definitions of the protection quantities in the recommendations of the International Commission on Radiological Protection set out in 2007 (Ann. ICRP37). It is expected that this change in operational quantities will impact current dosimeter designs. Although for many photon energies, the conversion coefficients from physical field quantities to the new operational quantities will change relatively little, for radiation fields with low energy photon components, such as medical x-ray applications, there will be a significant decrease in the values of the conversion coefficients. This means that the numerical values of the new operational quantities will be much lower for the same radiation field. These values will be closer to the effective dose, but this change can still cause confusion for medical staff. It is important to examine the effect of the new set of dose conversion coefficients on the personal dose in realistic radiation fields. We performed a study to assess the effect of changing the definition of the operational quantity, personal dose equivalent (Hp), in realistic radiation fields in interventional radiology (IR) workplaces. The x-ray tube kilovoltage peak (kVp) in IR ranges between 60 and 120 kV. The medical staff is exposed to the scattered photons which have a wide range of energies depending on the beam configuration and the patient size. The objective of this study is to 'quantitatively' estimate the impact of implementing the new ICRU quantities of Report 95 in IR radiation fields using Monte Carlo simulations. Simulations of 560 different configurations in IR were performed using MCNPX to calculate fluence binned per energy and angle of incidence.HpandHp(10)were then calculated for each configuration using dose conversion coefficients from fluence given by ICRU Reports 95 and 57, respectively. The results show that the mean of the ratio,Hp(10)/Hp, is 1.6 for all simulated scenarios. This reduction will correct the current overestimation of the effective dose and should result in better compliance with the dose limits in IR. However, it may also have negative consequences on the safety culture among the medical staff. Special care will be needed when interpreting these lower doses.

A parametric study of occupational radiation dose in interventional radiology by Monte-Carlo simulations.

This paper presents the results of a parametric study on the occupational exposure in interventional radiology to explore the influence of various variables on the staff doses. These variables include the angiography beam settings: x-ray peak voltage (kVp), added copper filtration, field diameter, beam projection and source to detector distance. The study was performed using Monte-Carlo simulations with MCNPX for more than 5600 combinations of parameters that account for different clinical situations. Additionally, the analysis of the results was performed using both multiple and random forest regression to build a predictive model and to quantify the importance of each variable when the variables simultaneously change. Primary and secondary projections were found to have the most effect on the scatter fraction that reaches the operator followed by the effect of changing the x-ray beam quality. The effect of changing the source to image intensifier distance had the lowest effect.

Development and Validation of the Realistic Anthropomorphic Flexible (RAF) Phantom.

Voxel phantoms developed by segmenting computed tomography images are known to be more anatomically accurate than mathematical phantoms. However, due to their lack of flexibility and the complexity of voxel datasets, the use of voxel phantoms in dosimetry is often impractical. This paper describes the development of the realistic anthropomorphic flexible (RAF) polygonal mesh phantom, a novel phantom based on Boundary Representation (B-Rep) that merges anatomical accuracy and flexibility. Rather than using segmentation of tomography images, the modeling of the phantom's organs was based on freely and commercially available anatomical atlases, ICRP 89, and recent medical literature. To validate the phantom, a high-resolution voxel version was created for the MCNPx transport code. The voxelized RAF phantom was validated by comparing it with the ICRP 110 male phantom for external irradiations with parallel beams of photons and electrons. Dose coefficients obtained from simulations with the RAF phantom were compared with those from ICRP Publication 116. Relevant differences in organ doses were found.

CREATION OF FEMALE COMPUTATIONAL PHANTOMS FOR CALIBRATION OF LUNG COUNTERS.

Plutonium isotopes are of high concern because they lead to high doses. In case of contamination, the activity burden inside the lungs should be assessed accurately. Many studies showed that the presence of breasts has a substantial influence on lung counting efficiencies. Currently, the calibration of most lung counting systems is done by means of physical phantoms representing only male chests. A set of female computational phantoms has been developed in order to provide gender-specific efficiency calibrations for the (241)Am gamma emission (59.54 keV). The phantoms were created starting from a library of female chest phantoms provided by Institut de radioprotection et de sûreté nucléaire (IRSN) (Farah, J. Amélioration des mesures anthroporadiamétriques personnalisées assistées par calcul Monte Carlo: optimisation des temps de calculs et méthodologie de mesure pour l'établissement de la répartition d'activite. PhD Thesis, 2011). While the IRSN phantoms represent a supine measurement position, the SCK•CEN lung counter set-up requires the persons to be sitting in a chair. Using open-source software, the breast shapes of the original phantoms have been recreated to simulate the drooping of breasts in vertical sitting position. A Monte Carlo approach was chosen for calculating calibration coefficients for female lung counting. The results obtained with MCNPx 2.7 simulations showed a significant decrease in the detection efficiency. For bigger bust and breast sizes, the detection efficiency showed to be up to 10 times lower than the ones measured with the Livermore male torso phantom.