A matheuristic for the selection of beam directions and dose distribution in Radiotherapy Planning

Mini EURO Conference on Improving Healthcare: new challenges, new approaches

物理学;医药卫生

Obal, T.M.^1,2 ; Florentino, H.O.^3 ; Gevert, V.G.^1 ; Jones, D.F.^4 ; Ouelhadj, D.^4 ; Volpi, N.M.P.^1 ; Wilhelm, V.E.^1

Federal University of Paraná, Curitiba/PR, Brazil^1

Federal Technological University of Paraná, Guarapuava/PR, Brazil^2

Biosciences Institute, UNESP, Botucatu/SP, Brazil^3

CORL, University of Porstmouth, Portsmouth, United Kingdom^4

关键词: Beam direction;    Computation time;    Dose distributions;    Intensity-modulated radiotherapy;    Interior-point method;    Optimisation models;    Optimum solution;    Radiotherapy planning;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/616/1/012012/pdf
DOI  :  10.1088/1742-6596/616/1/012012

学科分类：卫生学

来源: IOP

【 摘 要 】

In this paper a matheuristic using a combined Genetic Algorithm (GA) and exact method approach is proposed for selecting the position of the beams and dose distribution in Intensity Modulated Radiotherapy Planning (IMRT). GA selects a set of beams, for which the dose distribution is determined in the process of the GA's evaluation, using an optimisation model that is solved by an Interior Point method. Two instances are used to evaluate the performance of the matheuristic, comparing to the optimum solution, in terms of solution and computation time, found using the exact methodology of Branch and Bound. The results show that the matheuristic is appropriate to this problem in the case study proposed, as it is extremely faster than the exact method and also have reached the optimum solution in several experiments done.