ESTRO 2021 Abstract Book

S707

ESTRO 2021

In general, CK plans had lower GI than VMAT and Tomo plans and this GI advantage decreased with increasing target size. Among the CK techniques, CK Fixed had the lowest GI for 1cm target. For 1.5cm and 3cm targets, CK Fixed and CK Iris had comparable GI and which were lower than CK MLC. The GI advantage of CK Fixed and CK Iris over CK MLC decreased with increasing target size. For 5 cm target, CK MLC had both advantages of lower GI and shorter beam-on time. The distance of 70% IL/50% IL from the target provided a reference for expecting the maximum dose of a nearby organ. For example, for a 3cm spherical target using CK Fixed technique, maximum dose of an organ 0.3cm away from the target will be 70% of prescribed dose and maximum dose of an organ 0.55cm away from target will be 50% of prescribed dose. Conclusion CK had superior dose falloff compared to VMAT and Tomo. This advantage is most notable in the 1cm and 1.5cm target and least in the 5cm target. This suggested the dosimetric advantage of CK over VMAT and Tomo reduced as target size increases.Target size should be one of the considerations to selecting which CK techniques. CK Fixed seemed suitable for target size with diameter ≤ 1.5cm. CK Iris was suitable for target sizes between 1.5cm and 3cm. For target diameter ≥ 5cm, CK MLC might be the best choice.

In selecting treatment techniques, more factors other than dosimetry should be also considered, like treatment time, setup accuracy, imaging techniques, etc. For example, for a 3cm sphere target and patient in pain, CK MLC could be a better choice than CK Iris which sacrificed dosimetry slightly to reduce 24 mins beam- on time. PD-0871 A Quantum-Inspired Tensor Network for Optimizing Radiotherapy Plans S. Cavinato 1 , T. Felser 2 , M. Fusella 1 , M. Paiusco 1 , S. Montangero 2 1 Istituto Oncologico Veneto IOV-IRCCS, UOC Fisica Sanitaria, Padova, Italy; 2 Università degli Studi di Padova, Dipartimento di Fisica G. Galilei, Padova, Italy Purpose or Objective A new revolution in computational science has begun: the development of quantum computers. These machines will allow solving efficiently a wide class of highly complex computational problems. Quantum computing can be of great use in precision medicine, in particular in radiation therapy where the personalization of the treatment should take into acounto several elements, including the dose optimization. However, the lack of large-scale quantum computers puts hard constraints on what is currently feasible in practice. To partially overcome these limitations several quantum algorithms have been developed with the aim of both emulating and testing the behviour of quantum hardware on traditional computers. Among them, one the most effective solutions is represented by the so-called Tensor Network Methods (TNMs). Hereafter, an original feasibility study about the application of TNMs to IMRT beamlet intensity optimization is presented.

Materials and Methods