Development and Validation of Online Personal Dosimetry Application Using Computational Method for Interventional Cardiology

Introduction
Interventional cardiologists are often occupationally exposed to low radiation doses which put them at risk of stochastic radiation induced detriments. Therefore, individual monitoring of medical staff is essential to follow up regulatory dose limits and to apply the ALARA principle. However, current personal dosimeters are subject to large uncertainties, especially in non-homogeneous fields, like those found interventional radiology/cardiology. In these workplaces, medical staff should wear several dosimeters (extremity, eye lens, above/below apron) for a proper monitoring. However, the use of multiple dosimeters is unpractical, and in some cases it could hinder the work of the physicians (like in the case of finger dosimeters). As the capabilities of computational methods are increasing exponentially, it will become feasible to use pure computations to calculate doses in place of physical dosimeters. With this work, we present the current state of development of an innovative tool for calculating occupational doses using Monte-Carlo methods. The system is being developed within the PODIUM (Personal Online DosImetry Using computational Methods) research project.
Materials & Methods
In typical interventional radiology/cardiology scenarios, operators are exposed to non-homogeneous scatter radiation field coming from the body of the patient. The anisotropy is higher while working close by to the patient for performing manipulations. The two main inputs to our computational dosimetry system are: a) the spatiotemporal distribution of the scattered radiation field, including its intensity, its energy and its angular distributions; b) the relative position and pose of the operator in the scatter field.
1. Radiation field parameters. The scatter radiation is dependent on a number of factors such as: primary beam intensity, beam projection angle and patient thickness. Acquiring information about the primary beam and the patient can help reproducing the scatter field computationally. Imaging parameters includes kVp, filtration, collimation and beam projection are used to simulate the primary beam and its scattered field in Monte-Carlo simulations. At this stage, such information is obtained from a summary dose report after each procedure. The measured dose-area product (DAP) value allows to normalize the simulated relative doses (eV/g per particle) to the equivalent absolute dose units.
2. Operator motion tracking. The main input to compute doses to operators is the position and pose of the body of the operator relative to the X-ray beam and to the patient. Our system provides an indoor tracking system for tracking the position and the posture of the workers. The system is constituted by a Microsoft Kinect v2 Time-Of-Flight (TOF) camera and by an acquisition software package. The body skeleton information provided by the tracking system is then used to position a phantom.
At the current stage, the system represents a proof-of-concept and calculations are done off-line, after the procedure is finished, and the parameters of the procedure are collected. For validating our system, we performed tests in two interventional radiology (IR) rooms. In total, we followed 15 procedures in Cath-labs at UZ-VUB and CHU- ULiège hospitals. The Monte Carlo N-particle code (MCNPX 2.7) code [1] is used in our method for modelling and dosimetry calculations. The body skeleton information of the main operator provided by the tracking system is used to estimate the position of a dosimeter on the chest level in the simulations.
Results
An accurate analysis of the staff position was performed, and as a first step, we compared simulated Hp(10) with MCNP and measured Hp(10) with electronic personal dosimeter (EPD) Mk2.3 from Thermo Fisher Scientific worn above the lead apron during an angiography procedure for some of these procedures. The results showed good agreement with less than 5% difference between the calculated doses and the ones measured by the EPD dosimeter. The differences found in our simulations are easily explained by the uncertainties of the EPD dosemeter. In fact, the study performed by Clairand et al. [2] showed that the EPD Mk2.3 has a variation on the response within 30-40% due to the energy and angular response with the effect of the pulse frequency of the x-ray beam in interventional radiology fields.
In addition, simulations provided extra information about the eye lens dose Hp(3) to the operator during one procedure which shows the high spread of the ratio Hp(3)/Hp(10) between 0.48 to 1.75 for different beam projections due to field inhomogeneity.
Conclusions and future work
With this work, we show that simulating worker doses based on tracking systems and flexible phantoms in Monte-Carlo codes is possible. This method has big advantages in interventional radiology workplaces where the radiation fields are non-homogeneous and doses to staff can be relatively high. This method can also help for the application of the ALARA principle and for education and training of medical staff. For the future, we will transfer the skeletal data to the Realistic Anthropomorphic Flexible computational phantom [3] in Monte-Carlo simulation to calculate organ doses.
References
[1] D.B. Pelowitz, Ed., "MCNPX User’s Manual Version 2.7.0" LA-CP-11-00438 (2011).
[2] Clairand et al. “Use of active personal dosemeters in interventional radiology and cardiology: Tests in laboratory conditions and recommendations - ORAMED project”, Radiation Measurements, Volume 46, Issue 11, (2011).
[3] Lombardo et al. “Development and validation of the realistic anthropomorphic flexible (RAF) phantom”, Health Physics, 114:489–499, 05 (2018).
Acknowledgements
This project is funded by the CONCERT - European Joint Programme for the Integration of Radiation Protection Research 2014-2018 under grant agreement No. 662287.