Multi-parametric optimization of magnetic resonance imaging sequences for magnetic resonance-guided radiotherapy

• DOI: 10.1016/j.phro.2023.100497
• Publication Date: 2023-10-02T23:31:51Z
• Authors: Hafiz Muhammad Fahad; Stefan Dorsch; Moritz Zaiss; Christian P. Karger
• Keywords: CMA-ES; Pulse sequence

Background and PurposeMagnetic Resonance Imaging (MRI) is widely used in oncology for tumor staging, treatment response assessment, radiation therapy (RT) planning. This study proposes a framework automatic optimization of MRI sequences based on pulse sequence parameter sets (SPS) that are directly applied the scanner, application RT planning.Materials MethodsA phantom with 7 in-house fabricated contrasts was measurements. The proposed employed derivative-free algorithm to repeatedly update execute parametrized MR scanner acquire new data. In each iteration, mean-square error calculated clinical application. Two clinically relevant goals were pursued: achieving same signal therefore contrast as target image, maximizing difference (contrast) between specified tissue types. evaluated using two methods: covariance matrix adaptation evolution strategy (CMA-ES) genetic (GA).ResultsThe obtained results demonstrated potential sequences. Both CMA-ES GA methods showed promising goals, however, converged much faster compared GA.ConclusionsThe enables SPS it may be enhance quality images dedicated applications MR-guided RT.