Search

International audience; In order to address the recent concerns over a possible increasing in brain tumour mortality among interventional radiologists and cardiologist, this work evaluated the exposure conditions of the operator’s brain during interventional procedures using Monte Carlo simulations with anthropomorphic phantoms. The absorbed doses in several predefined segments of the operator’s brain were estimated in a typical interventional radiology irradiation scenario. The doses were normalized to the KAP values simulated for ten X-ray beam qualities and four projections (PA, RAO 25°, LAO 25° and CRA 25°). For the interventional radiology scenario, because of the position of the operator, no difference was found in the exposure between the left and right regions of the brain for the first operator, while for the second operator standing at a farer distance from the tube, the exposure of the left part of the brain is up to two times higher than that of the right part. The results are in agreement with dose measurements reported in the literature. The conversion factors, obtained as the absorbed dose per KAP, can be used to obtain a first estimate of the exposure of the brain of the operators during interventional procedures

International audience; Full Waveform Inversion (FWI) is a high-resolution seismic imaging tool, based on an iterative data fitting procedure. In this research, we focus on a near-surface application using three-component sources and three-component receivers (9C). The target is the Ettlingen Line, a defensive trench line located at Rheinstetten, Germany. In this work, we describe the first application of 3D elastic FWI method to a dense 3D dataset equipped with 9C. In practice, due to the limited number of equipment, the acquisition has been split into six parts (each part has the all source locations but only part of receiver locations). This separation leads to an inconsistent dataset concerning both amplitudes and phases. Therefore, a first step has led to the application of matching filter to homogenize the dataset. Several FWI strategies such as Vp - Vs parameter binding, gradient Bessel smoothing, and multi-scale approach have been considered during the inversion process to ensure a good convergence. Starting from a homogeneous model, we can achieve significant improvement in data-fit as well as a realistic reconstructed model. The location and dimension of the trench match with the previous experiments based on the inversion of surface wave dispersion curves with an additional increase in resolution.