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

Helge Egil Seime Pettersen*, Max Aehle, Johan Alme, Gergely Gábor Barnaföldi, Vyacheslav Borshchov, Anthony van den Brink, Mamdouh Chaar, Viljar Eikeland, Grigory Feofilov, Christoph Garth, Nicolas R. Gauger, Georgi Genov, Ola Grøttvik, Håvard Helstrup, Sergey Igolkin, Ralf Keidel, Chinorat Kobdaj, Tobias Kortus, Viktor Leonhardt, Shruti MehendaleRaju Ningappa Mulawade, Odd Harald Odland, Gábor Papp, Thomas Peitzmann, Pierluigi Piersimoni, Maksym Protsenko, Attiq Ur Rehman, Matthias Richter, Joshua Santana, Alexander Schilling, Joao Seco, Arnon Songmoolnak, Jarle Rambo Sølie, Ganesh Tambave, Ihor Tymchuk, Kjetil Ullaland, Monika Varga-Kofarago, Lennart Volz, Boris Wagner, Steffen Wendzel, Alexander Wiebel, Ren Zheng Xiao, Shiming Yang, Hiroki Yokoyama, Sebastian Zillien, Dieter Röhrich

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

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.

Original languageEnglish
Pages (from-to)1413-1418
Number of pages6
JournalActa Oncologica
Volume60
Issue number11
DOIs
Publication statusPublished - 14 Jul 2021

Keywords

  • convolutional neural network
  • machine learning
  • Monte Carlo simulation
  • Proton computed tomography
  • secondary particles
  • track reconstruction

Fingerprint

Dive into the research topics of 'Investigating particle track topology for range telescopes in particle radiography using convolutional neural networks'. Together they form a unique fingerprint.

Cite this