AUTHOR=Ma Pan , Xu Yingjie , Yao Yuhe , Lu Ningning , Dai Jianrong 

TITLE=Optimization method for determining vertices in lattice radiotherapy

JOURNAL=Frontiers in Oncology

VOLUME=Volume 15 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2025.1582402

DOI=10.3389/fonc.2025.1582402

ISSN=2234-943X

ABSTRACT=PurposeThis study presents an optimization method for arranging lattice radiotherapy (LRT) targets to enhance the contrast between peak and valley doses, aiming to improve the treatment effectiveness and precision.Materials and methodsThe LRT target comprises multiple sphere-like vertices generated using the optimization method, which involves four steps: 1) generating a volume for vertex arrangement, 2) determining initial positions and size of packing units, 3) determining initial positions and size of all the vertices and 4) optimizing the final vertex positions by using adaptive simulated annealing (ASA). Volumetric modulated arc therapy plans were retrospectively regenerated using the initial vertices produced by closest packing (Plan_Clo) and vertices obtained after ASA optimization (Plan_Opt). The peak-to-valley index (PVI) that characterizes the difference between peak and valley doses was introduced to evaluate the performance.ResultsA statistically significant difference was observed in the average PVI between Plan_Clo and Plan_Opt (p = 0.000). The average PVI ratio for Plan_Opt compared to Plan_Clo was 5.95 ± 4.87 (range: 1.24–16.80).ConclusionThe proposed optimization method for determining LRT target vertices has been validated, demonstrating a significant improvement in the PVI. ASA optimization, combined with closest packing, effectively enhanced the peak-to-valley dose difference in LRT, showcasing its potential for advancing treatment planning.