Treffer: RadField3D: a data generator and data format for deep learning in radiation-protection dosimetry for medical applications.

Title:
RadField3D: a data generator and data format for deep learning in radiation-protection dosimetry for medical applications.
Authors:
Lehner F; Physikalisch-Technische Bundesanstalt (PTB), Braunschweig, Germany.; Institute for Computer Graphics, Technische Universität Braunschweig, Braunschweig, Germany., Lombardo P; Belgian Nuclear Research Centre (SCK CEN), Boeretang, Mol, Belgium., Castillo S; Institute for Computer Graphics, Technische Universität Braunschweig, Braunschweig, Germany.; Cluster of Excellence PhoenixD, Leibniz University Hannover, Hannover, Germany., Hupe O; Physikalisch-Technische Bundesanstalt (PTB), Braunschweig, Germany., Magnor M; Institute for Computer Graphics, Technische Universität Braunschweig, Braunschweig, Germany.; Physics and Astronomy, University of New Mexico, Albuquerque, NM, United States of America.; Cluster of Excellence PhoenixD, Leibniz University Hannover, Hannover, Germany.
Source:
Journal of radiological protection : official journal of the Society for Radiological Protection [J Radiol Prot] 2025 May 16; Vol. 45 (2). Date of Electronic Publication: 2025 May 16.
Publication Type:
Journal Article; Research Support, Non-U.S. Gov't
Language:
English
Journal Info:
Publisher: IOP Pub. Ltd Country of Publication: England NLM ID: 8809257 Publication Model: Electronic Cited Medium: Internet ISSN: 1361-6498 (Electronic) Linking ISSN: 09524746 NLM ISO Abbreviation: J Radiol Prot Subsets: MEDLINE
Imprint Name(s):
Original Publication: [Bristol, UK] : IOP Pub. Ltd., [c1988-
MeSH Terms:
Deep Learning* , Radiometry*/methods , Radiation Protection*/methods, Monte Carlo Method ; Humans
Contributed Indexing:
Keywords: Monte-Carlo simulation; computational dosimetry; data format; data generator; deep learning; radiation
Entry Date(s):
Date Created: 20250507 Date Completed: 20250516 Latest Revision: 20250618
Update Code:
20250619
DOI:
10.1088/1361-6498/add53d
PMID:
40334671
Database:
MEDLINE

Weitere Informationen

In this research work, we present our open-source Geant4-based Monte-Carlo simulation application, called RadField3D, for generating three-dimensional radiation field datasets for dosimetry. Accompanying, we introduce a fast, machine-interpretable data format with a Python application programming interface for easy integration into neural network research, that we call RadFiled3D. Both developments are intended to be used to research alternative radiation simulation methods using deep learning. All data used for our validation (measured and simulated), along with our source codes, are published in separate repositories.https://github.com/Centrasis/RadField3DSimulationhttps://github.com/Centrasis/RadFiled3D.
(Creative Commons Attribution license.)