Click here to

Session: Therapy Radiotherapy Planning and Delivery [Return to Session]

Improving the Delivery Efficiency of Respiratory-Gated Radiotherapy Via Model-Predictive Control

E Salari1*, T Mazur2, G Sharp3, (1) Wichita State University, Wichita, KS, (2) Washington University School of Medicine, St. Louis, MO, (3) Massachusetts General Hospital, Boston, MA


WE-C1000-IePD-F7-4 (Wednesday, 7/13/2022) 10:00 AM - 10:30 AM [Eastern Time (GMT-4)]

Exhibit Hall | Forum 7

Purpose: Respiratory-gated radiotherapy is a common technique to manage intra-fraction motion; however, gating requires prolonged treatment delivery time. In this study, we demonstrate proof-of-principle for intra-fractional feedback of IMRT delivery via model-predictive control to achieve dosimetry comparable to respiratory gating without compromising delivery efficiency.

Methods: We develop a model-predictive control (MPC) approach based on pre-treatment 4DCT and an intra-fractional respiratory surrogate, where a cumulative dose model is used to continuously estimate the delivered dose distribution during irradiation. Step-and-shoot IMRT plans are independently optimized on each 4DCT phase for coverage of the per-phase GTV plus setup margin. Cumulative dose for different delivery regimens is estimated following delivery of each aperture via deformable image registration applied between phases and an average dataset. Given the cumulative dose delivered at a specified interval, an online (re-)optimization problem is solved to adjust the aperture intensities among the per-phase pool for the remainder of the treatment session. The subsequent aperture will be delivered and then re-optimization will be performed again in subsequent steps.

Results: Proof-of-principle simulation of the proposed MPC approach is performed using the 4D Extended Cardiac-Torso (XCAT) computational phantom while varying breathing amplitude, tumor size, and location for a lung tumor. Plan optimization and dose calculation were performed using an open-source treatment planning system (matRad). Specifically, we compare cumulative dose between the MPC approach and 1) ITV-based and 2) respiratory-gated treatments. The results show that the proposed method reduces irradiated tissue volume compared to ITV treatment while significantly reducing treatment time compared to traditional respiratory-gated treatment without compromising the target coverage.

Conclusion: We demonstrated proof-of-principle of a stochastic control approach for step-and-shoot IMRT delivery systems to continuously monitor the delivered dose and adjust the treatment plan accordingly to correct for dose discrepancies that may occur due to respiration-induced motion.

Funding Support, Disclosures, and Conflict of Interest: The research is funded in part by the National Science Foundation through Award No 1662819


Not Applicable / None Entered.


Not Applicable / None Entered.

Contact Email