Encrypted login | home

Program Information

Comparison of PlanIQ and RapidPlan Modeling for Radiotherapy Plan Quality Assurance

no image available
H Geng

H Geng1*, H Zhong1 , C Cheng1 , M Huang1 , S Ryu2 , Z Liao3 , F Yin4 , M Gillin3 , R Mohan3 , Y Xiao1 , (1)University of Pennsylvania, PA,(2) Stony Brook University Hospital, Stony Brook, NY, (3) UT MD Anderson Cancer Center, Houston, Texas, (4) Duke University Medical Center, Durham, NC


TU-C2-GePD-TT-1 (Tuesday, August 1, 2017) 10:00 AM - 10:30 AM Room: Therapy ePoster Theater

Purpose: To study the application of Plan IQ for radiation therapy (RT) treatment plan quality assurance (QA); And to compare Plan IQ and RapidPlan modeling for RT plan guidance.

Methods: RT plans submitted to RTOG 1308, RTOG 0631 were studied using Plan IQ. Six organ at risks (OAR) from twenty two plans in two trials were studied. Feasible OAR dose volume histogram (DVH) predicted by Plan IQ based on guaranteed target coverage were exported. RapidPlan models were built for RTOG 1308 and RTOG 0631, the OARs line objectives generated by the models were plotted against Plan IQ predictions. Their applications for plan quality assurance were analyzed.

Results: RTOG631 model prediction were all based on two full Arc geometry. RapidPlan line objective of SpinalCord_Prt from RTOG 631 plans predicts Dmax ranging from 8.1 to 10.4Gy, while plan IQ predicts achievable Dmax from 11.2Gy to 14.6 Gy indicating that PTV coverage were sacrificed in the library plans to preserve spinal cord. RapidPlan objectives for Cauda Equina from RTOG 631 plans predicts Dmax from 12.4 to 15.5Gy, while PlanIQ predicts 7.5 to 15Gy. This may suggest that better cauda equina sparing can be achieved when plan IQ prediction are used to guide planning. RTOG 1308 model prediction of Heart, lungs, esophagus and Spinal Cord line objectives all heavily depend on beam geometry and compared with planIQ predictions were only weakly affected by individual anatomy. PlanIQ may provide useful information for beam geometry QA.

Conclusion: Comparison of RapidPlan model generated line objectives with Plan IQ predictions can provide more insight on the quality of the library plans used in the models. Plna IQ gives information on ideal OAR DVHs, which may not be deliverable. While RapidPlan line objectives may provide achievable OAR DVHs heavily depend on beam geometry and clinical preferences of library plans.

Contact Email: