PLPasquale Lombardo

Pasquale Lombardo

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Pasquale Lombardo has a total of 15 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Learning-based estimation of spatially resolved scatter radiation fields in interventional radiology

    by Felix Timo Lehner, Pasquale Lombardo, Susana Castillo, Oliver Hupe, Marcus Magnor

    We present three variants of a lightweight, fully connected artificial neural network, suited for interactive estimation of three-dimensional, spatially resolved volumes of scattered radiation fields and a corresponding training pipeline for radiation protection dosimetry in medical radiation fields, such as those found in interventional radiology and cardiology. Accompanying, we present three different synthetically generated datasets with increasing complexity for training, generated using RadField3D, a Monte Carlo simulation application based on Geant4. As the primary scatter object, we employed the torso of a male Alderson RANDO phantom. On those datasets, we evaluate convolutional and fully connected architectures of neural networks to demonstrate which design decisions work well for reconstructing the fluence and spectra distributions over the spatial domain of such radiation fields. All our datasets, as well as our training pipeline, are published as open source in separate repositories. To evaluate the presented neural networks, we define and assess several metrics. Across these measures, the model variants demonstrate good spatial agreement between predicted and ground-truth radiation fields, particularly within specific regions of interest within the radiation field. Of particular relevance for potential application in out-of-field dosimetry is the symmetric mean absolute percentage error of the scatter radiation field, which represents the most challenging metric and was consistently above 84%.

  • Interactive posture program: a software platform for using the realistic anthropomorphic flexible phantom family in computational dosimetry applications

    by Pasquale Alessandro Lombardo, Dabin Jérémie, Mahmoud Abdelrahman, Daniel Santiago Rondon, Filip Vanhavere, Lara Struelens

    Boundary-representation (B-Rep) computational phantoms (CPs) offer high flexibility and anatomical accuracy, making them valuable tools in dosimetry and medical imaging simulations. However, their complexity and the need for expertise in graphic modelling limit their accessibility and adoption. Moreover, most Monte Carlo (MC) particle transport codes lack native support for B-Rep geometries, requiring complex and non-automated conversion to compatible formats, further complicating their use. To address these challenges, we developed the interactive posture program (IPP), a software platform to ease the use of B-Rep phantoms. The IPP features a real-time three-dimensional (3D) graphical interface, allowing users to easily create individualised phantoms and realistic exposure scenarios. Currently, the IPP includes three flexible phantoms—Adult Male, Adult Female, and 1-year-old Female—from the realistic anthropomorphic flexible (RAF) family, which can be extensively customised using integrated morphing tools to simulate diverse anatomical variations. Additionally, the IPP provides a comprehensive library of predefined exposure scenarios and advanced geometry creation tools, which allow the creation of accurate and individualised simulations. Finally, the IPP automatically generates simulation input files compatible with the most widely used MC transport codes, significantly simplifying the use of B-Rep phantoms within dosimetry and imaging simulations.


  • Benchmarking point-kernel method against Monte Carlo simulations for an ALARA case study in occupational radiation protection

    by Federica Roberto, Mahmoud Abdelrahman, Lombardo Pasquale, Filip Vanhavere, John A Lee

    The principle of as low as reasonably achievable (ALARA) is of paramount importance for the radiation protection of workers, and more specifically in the context of the decommissioning of nuclear facilities. For that purpose, several tools have been developed to measure and assess radiation doses to workers, to plan radiological work, and to ensure that safety protocols are followed. Most of these planning tools are based on the well-established point-kernel (PK) method, while others rely on Monte Carlo (MC) simulations. The scope of this paper is to compare, for some case studies, the performance and accuracy of VISIPLAN 4.0, a PK-based tool developed at SCK CEN, with MC simulations, to highlight the advantages and limitations of both methods. The major advantage of PK methods is usually related to their lower computational burden, at the cost of a less accurate assessment of dose values. However, depending on the number of source points chosen for the PK method, the simulation time could be comparable to MC simulations: in the scenarios presented in this paper, the simulation time was comparable to MC simulations when a high number of source points was considered. Moreover, VISIPLAN, on average, overestimated the dose assessment by approximately 50% compared to MC results.

  • Computational system for extremity dosimetry in nuclear medicine

    by Daniel Santiago Rondón, Pasquale Lombardo, Mahmoud Abdelrahman, Lara Struelens, Filip Vanhavere, Werner Schoonjans, Niki Bergans

    The goal of this research is to create a computational dosimetry system to monitor the dose received by nuclear medicine workers on their extremities. The system creates multiple Monte Carlo simulations to assess the dose received in different sections of the hands. Each simulation is created based on depth camera images of the manipulation of radioactive sources, using digital twins to represent the hands of the worker and the radiation source. The position of the hands and the source is constantly recalculated using computer vision algorithms. In this paper, we explain the different components of the computational system, assessing the possible sources of uncertainty introduced by each component. As a proof of concept, we compared the dose estimated by the computational system with real dosimeter measurements for two different scenarios. The simulated dose was, on average, 51% of the measured dose.