Encrypted login | home

Program Information

Automatic Treatment Planning for Whole Breast Radiation Therapy Using Hybrid IMRT

B Guo

B Guo*, C Shah , P Xia , The Cleveland Clinic Foundation, Cleveland, OH


SU-I-GPD-T-336 (Sunday, July 30, 2017) 3:00 PM - 6:00 PM Room: Exhibit Hall

Purpose: To introduce an automatic treatment planning technique for whole breast radiation therapy using hybrid IMRT.

Methods: MIM was used for atlas based segmentation of normal structures and identifying landmarks (wires and/or physician drawn boundaries) for beam placement. A MATLAB program was developed and integrated in MIM to setup tangential beams using the landmarks and generate Pinnacle Script for auto planning using hybrid IMRT. Energies of the beams were determined by the maximal separation and weightings of the beams were optimized using auto selected points within the breast. The hybrid IMRT plan contains a 3D plan prescribed to the maximum point dose and a step and shoot IMRT plan to compensate for the missing dose in the target while minimizing the doses to lungs and the heart. The 3D and IMRT plans were then merged to create a segmented tangential plan. The auto plans were compared with clinical plans for beam angles, MUs, target coverage, hot spots, and lung and heart doses for five patients.

Results: Auto planning produces similar beam setup with clinical plans: gantry and collimator angles were within 2 degrees. MUs of open beams were within 5%. MUs of segments were on average 4 times higher for auto plans. The increased modulation allowed for better coverage of the target (on average, PTV V95 increased from 92% to 95%, tumor bed D100 increased from 95% to 98%), less hot spots (PTV V105 decreased from 16% to 7%) and less lung and heart dose (ipsilateral Lung V20 reduced from 13% to 11% and heart mean dose reduced from 1.3 Gy to 1.1 Gy).

Conclusion: A fully automated treatment planning technique was developed for whole breast radiation therapy using hybrid IMRT. This technique gives equivalent or better plan quality compared with clinical plans.

Contact Email: