Research
Print page Print page
Switch language
The Capital Region of Denmark - a part of Copenhagen University Hospital
Published

Incorrect dosimetric leaf separation in IMRT and VMAT treatment planning: Clinical impact and correlation with pretreatment quality assurance

Research output: Contribution to journalJournal articleResearchpeer-review

  1. Dual-energy material decomposition for cone-beam computed tomography in image-guided radiotherapy

    Research output: Contribution to journalJournal articleResearchpeer-review

  2. A patch-based pseudo-CT approach for MRI-only radiotherapy in the pelvis

    Research output: Contribution to journalJournal articleResearchpeer-review

  3. Computed tomography synthesis from magnetic resonance images in the pelvis using multiple random forests and auto-context features

    Research output: Contribution to journalConference abstract in journalResearchpeer-review

  4. Patch-based generation of a pseudo CT from conventional MRI sequences for MRI-only radiotherapy of the brain

    Research output: Contribution to journalJournal articleResearchpeer-review

View graph of relations

PURPOSE: Dynamic treatment planning algorithms use a dosimetric leaf separation (DLS) parameter to model the multi-leaf collimator (MLC) characteristics. Here, we quantify the dosimetric impact of an incorrect DLS parameter and investigate whether common pretreatment quality assurance (QA) methods can detect this effect.

METHODS: 16 treatment plans with intensity modulated radiation therapy (IMRT) or volumetric modulated arc therapy (VMAT) technique for multiple treatment sites were calculated with a correct and incorrect setting of the DLS, corresponding to a MLC gap difference of 0.5mm. Pretreatment verification QA was performed with a bi-planar diode array phantom and the electronic portal imaging device (EPID). Measurements were compared to the correct and incorrect planned doses using gamma evaluation with both global (G) and local (L) normalization. Correlation, specificity and sensitivity between the dose volume histogram (DVH) points for the planning target volume (PTV) and the gamma passing rates were calculated.

RESULTS: The change in PTV and organs at risk DVH parameters were 0.4-4.1%. Good correlation (>0.83) between the PTVmean dose deviation and measured gamma passing rates was observed. Optimal gamma settings with 3%L/3mm (per beam and composite plan) and 3%G/2mm (composite plan) for the diode array phantom and 2%G/2mm (composite plan) for the EPID system were found. Global normalization and per beam ROC analysis of the diode array phantom showed an area under the curve <0.6.

CONCLUSIONS: A DLS error can worsen pretreatment QA using gamma analysis with reasonable credibility for the composite plan. A low detectability was demonstrated for a 3%G/3mm per beam gamma setting.

Original languageEnglish
JournalPhysica Medica
Volume32
Issue number7
Pages (from-to)918-25
Number of pages8
ISSN1120-1797
DOIs
Publication statusPublished - Jul 2016

    Research areas

  • Journal Article

ID: 49855282