title: A Comprehensive Framework for Dose Calculation in Intensity-Modulated Lung Cancer Particle Therapy
creator: Homolka, Noa Nikola
description: Lung cancer remains one of the leading causes of cancer-related mortality worldwide. Despite significant advancements in treatment options, lung cancer continues to present unique challenges, particularly in the context of particle therapy, including proton and carbon ion therapy. In addition to factors such as respiratory motion and the resulting dose calculation inaccuracies, tissue inhomogeneities can cause degradation of the treatment beam, potentially diminishing some of the advantages that particle therapy has over traditional X-rays. This thesis aims to quantify the impact of beam degradation on dose distributions for both analytical Pencil Beams (PBs) and Monte Carlo (MC) simulations, exploring how this degradation influences dose accuracy, conformity, and overall relevance compared to other error sources and uncertainties. The central question to be answered is whether this degradation is significant and whether it has the same impact on analytical PBs compared to MC simulations, particularly in the context of Relative Biological Effectiveness (RBE)-weighted dose calculation. To address these questions, this thesis developed a dose calculation module for the inclusion of dose degradation in both analytical and MC treatment plans within the open-source toolkit matRad with a specific focus on proton and carbon ion therapy for lung cancer treatment.  A MC interface was developed for the matRad toolkit as part of this thesis, originally designed to test and assess the impact of degradation effects on more realistic and accurate MC simulations. Dose distributions calculated through the interface were first validated on homogeneous water geometries with newly developed generic machine data sets based on MC simulations, that showed near-perfect agreement between MC engines and analytical PBs. This allows to focus on changes arising from differences in geometry, the currently studied effect, or the chosen algorithms. Due to its versatility and modular setup, it can not only be used for investigations in degradation correction, but it allows for applications in various research projects and has already seen extensive use within the department.  For analytical PBs, the existing dose degradation implementation was refined for RBE-weighted dose calculations for protons and carbon ions. A universal implementation of degradation correction, independent of the dose calculation algorithm, was realized through a density sampling model utilizing a simple beta distribution. The integration of these degradation models was explored and tested on increasingly complex geometries, starting from simple lung box phantoms and progressing to patient treatment plans. The degradation effects were consistent between carbon ions and protons, and in some patient cases, degradation effects for variable RBE-weighted proton doses were even able to improve dose conformity and homogeneity. The observed degradation effects on dose accuracy increased for more complex scenarios, but were overall minimal, even considering the worst-case modulation power.  Potentially significant increases in the dose to Organs At Risk (OARs) was noted, that did not lead to a substantial increase in Normal Tissue Complication Probability (NTCP). However, in certain cases, the impact of dose degradation can be significant, particularly when added on top of the already degraded MC simulations. It was found that other error sources, such as organ motion and comparison with more accurate MC simulations, most likely overshadow the impact of degradation effects. Evaluation of this aspect should certainly be part of future work.  Additionally, the variability in results was found to depend strongly on patient-specific factors such as tumor size and location within the lung. While for larger tumors or tumors with minimal lung tisse in the beam path, degradation effects may indeed be negligible, for small tumors located deep within lung tissue, the impact can be significant, particularly in cases involving carbon ions.  In conclusion, this thesis presents a comprehensive framework for dose calculation in intensity-modulated lung cancer particle therapy through the integration of a MC interface for matRad. This enables the calculation of degraded absorbed and RBE-weighted dose distributions for both protons and carbon ions, providing a valuable tool for advancing the accuracy and effectiveness of lung cancer treatment.
date: 2025
type: Dissertation
type: info:eu-repo/semantics/doctoralThesis
type: NonPeerReviewed
format: application/pdf
identifier: https://archiv.ub.uni-heidelberg.de/volltextserver/36588/1/Noa_Homolka_Dissertation.pdf
identifier: DOI:10.11588/heidok.00036588
identifier: urn:nbn:de:bsz:16-heidok-365889
identifier:   Homolka, Noa Nikola  (2025) A Comprehensive Framework for Dose Calculation in Intensity-Modulated Lung Cancer Particle Therapy.  [Dissertation]     
relation: https://archiv.ub.uni-heidelberg.de/volltextserver/36588/
rights: info:eu-repo/semantics/openAccess
rights: http://archiv.ub.uni-heidelberg.de/volltextserver/help/license_urhg.html
language: eng