Differences in radiotherapy delivery and outcome due to contouring variation

Publikation: Bidrag til bog/antologi/rapportKonferencebidrag i proceedingsForskningfagfællebedømt

Gross tumor volume (GTV) delineation is central for radiotherapy planning. It provides the basis of the clinical target volume and, ultimately, the planning target volume which is used for dose optimization. Manual GTV delineations are prone to intra- and inter-observer variation and automatic segmentation methods also produce different results. There is no consensus on how to account for the contouring uncertainty, but has been suggested to incorporate it into the planning target volume (PTV) margin. Current recipes for the PTV margin are based on normal distribution assumptions and are more suitable for setup and execution errors. In this study we use the GTV delineations made by 6 experienced clinicians to create delineation-specific dose plans. These dose plans are then used to calculate theoretic tumor control probabilities (TCP) differences between delineations. The results show that current margin recipes are inadequate for maintaining the same TCP despite manual delineation variation. New methods to account for delineation variation should be developed.

OriginalsprogEngelsk
TitelClinical Image-Based Procedures : From Planning to Intervention - International Workshop, CLIP 2012, Held in Conjunction with MICCAI 2012, Revised Selected Papers
Antal sider8
ForlagSpringer Verlag
Publikationsdato2013
Sider122-129
ISBN (Trykt)9783642380785
DOI
StatusUdgivet - 2013
BegivenhedInternational Workshop on Clinical Image-Based Procedures: From Planning to Intervention, CLIP 2012 - Held in Conjunction with MICCAI 2012 - Nice, Frankrig
Varighed: 5 okt. 20125 okt. 2012

Konference

KonferenceInternational Workshop on Clinical Image-Based Procedures: From Planning to Intervention, CLIP 2012 - Held in Conjunction with MICCAI 2012
LandFrankrig
ByNice
Periode05/10/201205/10/2012
Sponsorexocad GmbH, MedCom GmbH
NavnLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Vol/bind7761 LNCS
ISSN0302-9743

ID: 357363967