Back to search

Georgia Institute of Technology

Brain Necrosis Prediction Using a Dose-Let Volume Histogram (DLVH)

Abstract

dc:description.abstract

Purpose: A novel tool of dose linear-energy-transfer (LET) volume histogram (DLVH) was utilized to derive dose-LET volume constraints (DLVCs) for brain necrosis (BN) in central nervous system (CNS) cancer patients treated with proton therapy by comparing DLVHs between the control group and adverse event (AE) group. Methods: A retrospective analysis of an institutional outcomes database was performed to find adult CNS cancer patients undergoing proton therapy who developed BN after treatment, forming the AE group. Patients who did not develop BN post-treatment were matched and paired with the AE patients by radiation dose, fractionation, tumor stage, and BN grade. Dose and LET distributions were calculated by an in-house developed Monte Carlo dose engine, which were used to construct DLVHs of the brain for all patients. The DLVH indices, V(d Gy, l keV/μm), which measures the percentage of the volume in the selected structure that has a dose at least d Gy and a LET at least l keV/μm, were compared in the dose-LET plane between the control and AE group and a p-value map in the dose-LET plane was then derived using the Mann-Whitney U Test. DLVCs were then derived from statistical significant features found in the p-value map to differentiate the control and AE group. Reproducibility was evaluated by calculating the concordance correlation coefficient (CCC) voxel-wise throughout the examined dose-LET space. Results: 17 patients were analyzed; 7 AE patients matched to 10 control patients. Two DLVCs were derived from these regions: DLVC1 is V(56.5Gy, 0keV/μm), representing a high dose constraint (p=0.047), and DLVC2 is V(43.5Gy, 4.25keV/μm), representing a high LET constraint (p=0.039). Conclusion: DLVH was successfully used to deduce DLVCs for CNS cancer patients undergoing proton therapy. The derived DLVCs can be used in the proton therapy plan evaluation retrospectively and in the treatment planning prospectively to reduce possible BNs. This study demonstrates evidence of the LET-enhancing effect for the AE initialization in proton therapy.

Degree

thesis:*
Level thesis:degree_level
Masters
Department dc:contributor.department
Mechanical Engineering
Grantor dc:publisher
Georgia Institute of Technology
Year dc:date.issued
2024

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Yu, Justin
Advisors dc:contributor.advisor
  • Wang, C.-K. Chris
  • Liu, Wei
Committee members dc:contributor.committeemember
  • Bues, Martin
  • Flampouri, Stella

Subjects

dc:subject × 3

Identifiers

dc:identifier.*
Handle dc:identifier.uri
https://hdl.handle.net/1853/75624

Chain of custody

source
Harvested from
Georgia Tech
Base URL
repository.gatech.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Yu, Justin. Brain Necrosis Prediction Using a Dose-Let Volume Histogram (DLVH). Masters thesis, Georgia Institute of Technology, 2024. https://hdl.handle.net/1853/75624