TSchainRad: a new Tabu Search-Based Matheuristic for IMRT Optimization
Luis Tertulino, Sílvia Maria Diniz Monteiro Maia, Elizabeth Ferreira Gouvêa Goldbarg, Marco C. Goldbarg · 2019
Radiotherapy is a form of treatment that employs ionizing radiation to deal with cancerous tissues. It is desired to administer a dose of radiation to the tumor region sufficient to destroy it while sparing the healthy anatomical structures. Optimization techniques are used to create plans that satisfy that goal. This paper presents a new Tabu Search matheuristic coupled with an ejection chain procedure for the geometry problem that arises from Intensity Modulated Radiotherapy plans. A random restart provides diversification. The method proposed in this study was applied to ten real liver cancer cases, which are considered hard instances due to the number of surrounding healthy structures. The results with those produced by a Tabu Search method from the literature are compared. The analysis of those results shows that the proposed approach is promising since it provides overall better plans.