TY - JOUR
T1 - Lung 4D-IMRT treatment planning
T2 - An evaluation of three methods applied to four-dimensional data sets
AU - Ehler, Eric D.
AU - Tomé, Wolfgang A.
PY - 2008/9
Y1 - 2008/9
N2 - Purpose: To compare 4D-dose distributions for IMRT planning on three data sets: a single 4D-CT phase, a 4D-CT phase with a density override to the tumor motion envelope (TME) volume, and the average intensity projection (AIP). Methods: Eight planning cases were considered. IMRT inverse planning optimization was performed on each of the three data set types, for each case considered. The plans were then applied to all ten phases of the associated 4D-CT data set. The dose to the GTV in each breathing phase was compared to the TME dose from the optimized dose distribution, as well as the GTV dose determined from a model-based deformable registration algorithm. Results: IMRT optimization on a single 3D data set resulted in a greater equivalent uniform dose (EUD) to the GTV when applied to a 4D-CT data set than the EUD for the TME in the optimized plan. The difference was up to 5.5 Gy in one case. For all cases and planning techniques considered, a maximum difference of 0.3 Gy in the NTDmean to the healthy lung throughout the breathing cycle was found. Conclusions: For tumors located in the periphery of the lung, optimization on the AIP image resulted in a more uniform GTV dose throughout the breathing cycle. Averages in GTV EUD and healthy lung NTDmean taken over all the breathing phases were found to be in agreement with the dose effect parameters obtained from model-based deformable registration algorithms. All planning methods yielded GTV EUD values that were larger than the prescribed dose when the full 4D data set was considered.
AB - Purpose: To compare 4D-dose distributions for IMRT planning on three data sets: a single 4D-CT phase, a 4D-CT phase with a density override to the tumor motion envelope (TME) volume, and the average intensity projection (AIP). Methods: Eight planning cases were considered. IMRT inverse planning optimization was performed on each of the three data set types, for each case considered. The plans were then applied to all ten phases of the associated 4D-CT data set. The dose to the GTV in each breathing phase was compared to the TME dose from the optimized dose distribution, as well as the GTV dose determined from a model-based deformable registration algorithm. Results: IMRT optimization on a single 3D data set resulted in a greater equivalent uniform dose (EUD) to the GTV when applied to a 4D-CT data set than the EUD for the TME in the optimized plan. The difference was up to 5.5 Gy in one case. For all cases and planning techniques considered, a maximum difference of 0.3 Gy in the NTDmean to the healthy lung throughout the breathing cycle was found. Conclusions: For tumors located in the periphery of the lung, optimization on the AIP image resulted in a more uniform GTV dose throughout the breathing cycle. Averages in GTV EUD and healthy lung NTDmean taken over all the breathing phases were found to be in agreement with the dose effect parameters obtained from model-based deformable registration algorithms. All planning methods yielded GTV EUD values that were larger than the prescribed dose when the full 4D data set was considered.
KW - 4D IMRT
KW - 4D-CT
KW - Average intensity projection
KW - Intrafraction motion
KW - Tumor motion
UR - http://www.scopus.com/inward/record.url?scp=52049095720&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=52049095720&partnerID=8YFLogxK
U2 - 10.1016/j.radonc.2008.07.004
DO - 10.1016/j.radonc.2008.07.004
M3 - Article
C2 - 18703249
AN - SCOPUS:52049095720
SN - 0167-8140
VL - 88
SP - 319
EP - 325
JO - Radiotherapy and Oncology
JF - Radiotherapy and Oncology
IS - 3
ER -