Lux Ex Machina – an application of machine learning in glow curve deconvolution
Final Report Abstract
Personal dosimetry is an important aspect of radiation protection, in particular in medical fields such as radiology or nuclear medicine. It is also a relevant field of research as current technologies for dose monitoring, in particular film badge dosimeters, are pushed to their limits. Viable alternatives are e.g. based on the well-known phenomenon of thermoluminescence (TL). The TL-DOS dosimetry system, jointly developed by the Materialprüfungsamt NRW, Dortmund, and TU Dortmund University, uses a thin-layer of LiF:Mg,Ti as thermoluminescent material and fast heating procedures. The system can be used to reliably estimate irradiation doses by counting the number of photons emitted during the heating process and a proper calibration. The aim of the project was to maximize the information obtained from the time-resolved photon signal, the glow curve, by simulating the fundamental TL processes and by using modern and powerful machine learning algorithms. The information strived for was the irradiation dose, the time between irradiation and read-out, and the number of irradiations within one duty cycle. Based on simulation data and data obtained from well-defined irradiation campaigns with photons, the machine learning algorithms were trained, tested and validated, and the full potential of glow curve deconvolution has been explored.
Publications
-
Entwicklung eines „rAIdiation protection co pilot‘s", 95. AKD Sitzung, Peine, DE
Derugin, Evelin; Kröninger, Kevin; Mentzel, Florian; Nackenhorst, Olaf; Walbersloh, Jörg & Weingarten, Jens
-
Erweiterung der für die Überwachung der Personendosis relevanten Informationen aus den Glühkurven von Thermolumineszenzdosimetern unter Verwendung generativer neuronaler Netze, 51. Jahrestagung der DGMP 2020
Derugin, Evelin; Jansen, Hannah; Kröninger, Kevin; Rempe, Moritz; Walbersloh, Jörg & Weingarten, Jens
-
Improving information extracted from glow curves of thermoluminescence personal dosemeters using CNNs, DPG Spring Meeting 2021
Derugin, Evelin; Kröninger, Kevin; Mentzel, Florian; Walbersloh, Jörg & Weingarten, Jens
-
Improving the analysis of TL dosimeter with convolutional neural networks, Joint Conference of the ÖGMP, DGMP and SGSMP 2021
Derugin, Evelin; Jansen, Hannah; Kröninger, Kevin; Mentzel, Florian; Nackenhorst, Olaf; Walbersloh, Jörg & Weingarten, Jens
-
No more glowing in the dark: how deep learning improves exposure date estimation in thermoluminescence dosimetry. Journal of Radiological Protection, 41(4), S506-S521.
Mentzel, Florian; Derugin, Evelin; Jansen, Hannah; Kröninger, Kevin; Nackenhorst, Olaf; Walbersloh, Jörg & Weingarten, Jens
-
Analyse von TLD-Dosimetern mittels neuronaler Netzwerke, LPS: 22. Sommerschule im Strahlenschutz 2022, Berlin, DE
Derugin, Evelin
-
Deep TL: progress of a machine learning aided personal dose monitoring system, 6th European Congress on Radiation Protection 2022, Budapest, HU
Derugin, Evelin; Kröninger, Kevin; Mentzel, Florian; Nackenhorst, Olaf; Walbersloh, Jörg & Weingarten, Jens
-
Der „rAIdiation protection co-pilot“: Ein Machine Learning unterstütztes Bestrahlungsszenario-Vorhersage-Tool, 53. FS Jahrestagung 2022, Konstanz, DE
Derugin, Evelin; Kröninger, Kevin; Mentzel, Florian; Nackenhorst, Olaf; Walbersloh, Jörg & Weingarten, Jens
-
rAIdiation protection co-pilot: A Machine Learning aided radiation scenario prediction tool, 64th Annual Meeting & Exhibition AAPM 2022, Washington, DC, USA
Derugin, Evelin; Kröninger, Kevin; Mentzel, Florian; Nackenhorst, Olaf; Walbersloh, Jörg & Weingarten, Jens
-
Deep TL: progress of a machine learning aided personal dose monitoring system. Radiation Protection Dosimetry, 199(8-9), 767-774.
Derugin, Evelin; Kröninger, Kevin; Mentzel, Florian; Nackenhorst, Olaf; Walbersloh, Jörg & Weingarten, Jens
-
Thermoluminescence glow curve generation using generative adversarial networks (GANs), DPG Spring Meeting 2023, Dresden
Derugin, Evelin; Kröninger, Kevin; Mentzel, Florian; Nackenhorst, Olaf; Walbersloh, Jörg & Weingarten, Jens
