Feasibility of artificial-intelligence-based synthetic computed tomography in a magnetic resonance-only radiotherapy workflow for brain radiotherapy: Two-way dose validation and 2D/2D kV-image-based positioning

• DOI: 10.1016/j.phro.2022.10.002
• Publication Date: 2022-10-22T07:07:46Z
• Authors: Siti Masitho; Juliane Szkitsak; Johanna Grigo; Rainer Fietkau; Florian Putz; Christoph Bert
• Keywords: Image-guided radiation therapy

Magnetic Resonance Imaging (MRI)-only workflow eliminates the MRI-computed tomography (CT) registration inaccuracy, which degrades radiotherapy (RT) treatment accuracy. For an MRI-only MRI sequences need to be converted synthetic-CT (sCT). The purpose of this study was evaluate a commercially available artificial intelligence (AI)-based sCT generation for dose calculation and 2D/2D kV-image daily positioning brain RT workflow.T1-VIBE DIXON acquired at 1.5 T 26 patients in setup sCTs generation. each patient, volumetric modulated arc therapy (VMAT) plan optimized on CT, then recalculated sCT; vice versa. sCT-based digitally reconstructed radiographs (DRRs) were fused with stereoscopic X-ray images recorded as image guidance clinical treatments. Dosimetric differences between planned/recalculated doses calculated couch shift/rotation evaluated.Mean ΔD50 target volumes ranged -0.2 % 0.2 %; mean ΔD0.01ccm -0.6 1.6 -1.4 1.0 organ-at-risks, respectively. Differences tested equivalence using intervals ±2 (dose), ±1mm (translation), ±1° (rotation). Dose found interval (p < 0.001). median lat./long./vert. shift CT-based/sCT-based DRRs 0.3 mm/0.2 mm/0.3 mm 0.05); rotation -1.5°/0.1°/0.1° (after improvement setup: -0.4°/-0.1°/-0.4°, p 0.05).This in-silico showed that AI-based provided equivalent results CT when optimal during acquisition.