Investigating particle track topology for range telescopes in particle radiography using convolutional neural networks

Authors

  • Helge Egil Seime Pettersen Department of Oncology and Medical Physics, Haukeland University Hospital, Bergen, Norway
  • Max Aehle Chair for Scientific Computing, Technische Universität Kaiserslautern, Kaiserslautern, Germany
  • Johan Alme Department of Physics and Technology, University of Bergen, Bergen, Norway
  • Gergely Gábor Barnaföldi Wigner Research Centre for Physics, Budapest, Hungary
  • Vyacheslav Borshchov Research and Production Enterprise “LTU” (RPE LTU), Kharkiv, Ukraine
  • Anthony van den Brink Institute for Subatomic Physics, Utrecht University/Nikhef, Utrecht, Netherlands
  • Mamdouh Chaar Department of Physics and Technology, University of Bergen, Bergen, Norway
  • Viljar Eikeland Department of Physics and Technology, University of Bergen, Bergen, Norway
  • Grigory Feofilov Department of High Energy and Elementary Particles Physics, St. Petersburg University, St. Petersburg, Russia
  • Christoph Garth Scientific Visualization Lab, Technische Universität Kaiserslautern, Kaiserslautern, Germany
  • Nicolas R. Gauger Chair for Scientific Computing, Technische Universität Kaiserslautern, Kaiserslautern, Germany
  • Georgi Genov Department of Physics and Technology, University of Bergen, Bergen, Norway
  • Ola Grøttvik Department of Physics and Technology, University of Bergen, Bergen, Norway
  • Håvard Helstrup Department of Computer Science, Electrical Engineering and Mathematical Sciences, Western Norway University of Applied Sciences, Bergen, Norway
  • Sergey Igolkin Department of High Energy and Elementary Particles Physics, St. Petersburg University, St. Petersburg, Russia
  • Ralf Keidel Center for Technology and Transfer (ZTT), University of Applied Sciences Worms, Worms, Germany
  • Chinorat Kobdaj Institute of Science, Suranaree University of Technology, Nakhon Ratchasima, Thailand
  • Tobias Kortus Center for Technology and Transfer (ZTT), University of Applied Sciences Worms, Worms, Germany
  • Viktor Leonhardt Scientific Visualization Lab, Technische Universität Kaiserslautern, Kaiserslautern, Germany
  • Shruti Mehendale Department of Physics and Technology, University of Bergen, Bergen, Norway
  • Raju Ningappa Mulawade Center for Technology and Transfer (ZTT), University of Applied Sciences Worms, Worms, Germany
  • Odd Harald Odland Department of Oncology and Medical Physics, Haukeland University Hospital, Bergen, Norway; Department of Physics and Technology, University of Bergen, Bergen, Norway
  • Gábor Papp Institute for Physics, Eötvös Loránd University, Budapest, Hungary
  • Thomas Peitzmann Institute for Subatomic Physics, Utrecht University/Nikhef, Utrecht, Netherlands
  • Pierluigi Piersimoni Department of Physics and Technology, University of Bergen, Bergen, Norway
  • Maksym Protsenko Research and Production Enterprise “LTU” (RPE LTU), Kharkiv, Ukraine
  • Attiq Ur Rehman Department of Physics and Technology, University of Bergen, Bergen, Norway
  • Matthias Richter Department of Physics, University of Oslo, Oslo, Norway
  • Joshua Santana Center for Technology and Transfer (ZTT), University of Applied Sciences Worms, Worms, Germany
  • Alexander Schilling Center for Technology and Transfer (ZTT), University of Applied Sciences Worms, Worms, Germany
  • Joao Seco Department of Biomedical Physics in Radiation Oncology, DKFZ-German Cancer Research Center, Heidelberg, Germany; Department of Physics and Astronomy, Heidelberg University, Heidelberg, Germany
  • Arnon Songmoolnak Department of Physics and Technology, University of Bergen, Bergen, Norway; Institute of Science, Suranaree University of Technology, Nakhon Ratchasima, Thailand
  • Jarle Rambo Sølie Department of Diagnostic Physics, Division of Radiology and Nuclear Medicine, Oslo University Hospital, Oslo, Norway
  • Ganesh Tambave Department of Physics and Technology, University of Bergen, Bergen, Norway
  • Ihor Tymchuk Research and Production Enterprise “LTU” (RPE LTU), Kharkiv, Ukraine
  • Kjetil Ullaland Department of Physics and Technology, University of Bergen, Bergen, Norway
  • Monika Varga-Kofarago Wigner Research Centre for Physics, Budapest, Hungary
  • Lennart Volz Department of Biophysics, GSI Helmholtz Center for Heavy Ion Research GmbH, Darmstadt, Germany; Department of Medical Physics and Biomedical Engineering, University College London, London, UK
  • Boris Wagner Department of Physics and Technology, University of Bergen, Bergen, Norway
  • Steffen Wendzel Center for Technology and Transfer (ZTT), University of Applied Sciences Worms, Worms, Germany
  • Alexander Wiebel Center for Technology and Transfer (ZTT), University of Applied Sciences Worms, Worms, Germany
  • RenZheng Xiao Department of Physics and Technology, University of Bergen, Bergen, Norway; College of Mechanical & Power Engineering, China Three Gorges University, Yichang, People’s Republic of China
  • Shiming Yang Department of Physics and Technology, University of Bergen, Bergen, Norway
  • Hiroki Yokoyama Institute for Subatomic Physics, Utrecht University/Nikhef, Utrecht, Netherlands
  • Sebastian Zillien Center for Technology and Transfer (ZTT), University of Applied Sciences Worms, Worms, Germany
  • Dieter Röhrich Department of Physics and Technology, University of Bergen, Bergen, Norway

DOI:

https://doi.org/10.1080/0284186X.2021.1949037

Keywords:

Proton computed tomograph, machine learning, Monte Carlo simulation, track reconstruction, convolutional neural network, secondary particles

Abstract

Background

Proton computed tomography (pCT) and radiography (pRad) are proposed modalities for improved treatment plan accuracy and in situ treatment validation in proton therapy. The pCT system of the Bergen pCT collaboration is able to handle very high particle intensities by means of track reconstruction. However, incorrectly reconstructed and secondary tracks degrade the image quality. We have investigated whether a convolutional neural network (CNN)-based filter is able to improve the image quality.

Material and methods

The CNN was trained by simulation and reconstruction of tens of millions of proton and helium tracks. The CNN filter was then compared to simple energy loss threshold methods using the Area Under the Receiver Operating Characteristics curve (AUROC), and by comparing the image quality and Water Equivalent Path Length (WEPL) error of proton and helium radiographs filtered with the same methods.

Results

The CNN method led to a considerable improvement of the AUROC, from 74.3% to 97.5% with protons and from 94.2% to 99.5% with helium. The CNN filtering reduced the WEPL error in the helium radiograph from 1.03 mm to 0.93 mm while no improvement was seen in the CNN filtered pRads.

Conclusion

The CNN improved the filtering of proton and helium tracks. Only in the helium radiograph did this lead to improved image quality.

Downloads

Download data is not yet available.

Downloads

Additional Files

Published

2021-11-02

How to Cite

Egil Seime Pettersen, H., Aehle, M., Alme, J., Gábor Barnaföldi, G., Borshchov, V., van den Brink, A., … Röhrich, D. (2021). Investigating particle track topology for range telescopes in particle radiography using convolutional neural networks. Acta Oncologica, 60(11), 1413–1418. https://doi.org/10.1080/0284186X.2021.1949037

Issue

Vol. 60 No. 11 (2021): Acta Oncologica

Section

Articles