169. Genetic algorithm based script for automation of head and neck VMAT treatment planning. (December 2018)

• Record Type: Journal Article
• Authors: Fiandra, C.; Loi, G.; Fusella, M; Scaggion, A.; Alparone, A.; Giglioli, F.R.; Paiusco, M.
• Is Part Of: Physica medica. Volume 56(2018)Supplement 2
• Page Start: 168
• Publication Date: 2018-12
• Subjects: Medical physics -- Periodicals; Biophysics -- Periodicals; Biophysics -- Periodicals; Imagerie médicale -- Périodiques; Radiothérapie -- Périodiques; Rayons X -- Sécurité -- Mesures -- Périodiques; Physique -- Périodiques; Médecine -- Périodiques; 610.153
• DOI: 10.1016/j.ejmp.2018.04.180
• Languages: English
• ISSNs: 1120-1797

Abstract

Abstract: Purpose: To use a specific Phyton script for automatic optimization process in RayStation treatment planning system (TPS) in the head and neck treatments. Comparison with different commercial automatic planning is reported. Methods: Genetic algorithm (GA) was implemented for head and neck VMAT treatments. The chromosomes of the algorithm were max equivalent uniform dose functions (maxEUD) for: larynx, oral cavity and both parotid glands. Ten initial couples were randomly generated and a maximum of five reproductions were permitted. The script was tested for fifteen patients in three different centres: some patients had three dose levels (70 Gy, 63 Gy and 54 Gy) in 35 fractions, others had two levels (69.96 Gy, 60 Gy) in 32 fractions. A comparison with corresponding plans created with Monaco TPS (M, Elekta), MCO option of RayStation (MCO, RaySearch) and RapidPlan (RP, Varian) was carried out. All the plans were evaluated with a score of PlanIQ (Sun Nuclear) in terms of target coverage, conformity index of PTVs and constraints of OARs. Results: In figure are reported the totals scores (TS) obtained. Script plans scores were always higher than RP and M (mean TS: 80.0 ± 9.1 for GA and 72.2 ± 9.1 for RP, 77.3 ± 13.9 for GA and 74.2 ± 14.2 for M). PTVs coverage, Maximum doses of spinal cord and brainstem were always below the requested value for all tested systems. For parallel OARs, GA spared parotid glands more than RP (average score of one parotid gland: 81.0 ± 15.0 for GA and 45.0 ± 10.0 for RP); GA spared also Larynx more than MCO (average score 78.0 ± 9.0 for GA and 29.0 ± 12.0 for MCO). For other OARs, the results obtained with GA, RP and MCO were similar. Conclusions: GA gives results clinically comparable with automatic commercial systems. These preliminary results must be confirmed with a larger Head and Neck plans dataset[1] .