Evaluation of the influence of susceptibility-induced magnetic field distortions on the precision of contouring intracranial organs at risk for stereotactic radiosurgery. (July 2020)

• Record Type: Journal Article
• Authors: Mengling, Veit; Putz, Florian; Laun, Frederik Bernd; Perrin, Rosalind; Eisenhut, Felix; Dörfler, Arnd; Fietkau, Rainer; Bert, Christoph
• Is Part Of: Physics and imaging in radiation oncology. Volume 15(2020)
• Page Start: 91
• Page End: 97
• Publication Date: 2020-07
• Subjects: Susceptibility artifacts -- MRI -- Safety margins -- Stereotactic -- Brain -- PRV; Radiotherapy -- Periodicals; Radiation dosimetry -- Periodicals; Cancer -- Imaging -- Periodicals; Oncology -- Periodicals; 615.842
• DOI: 10.1016/j.phro.2020.08.001
• Languages: English
• ISSNs: 2405-6316

Abstract

Highlights: 45 data sets (18 on a 1.5 T MR and 27 on a 3 T MR) were evaluated for susceptibility induced distortions. Maximum distortions of up to 1.7 mm were found for organs at risk in standard diagnostic settings. Median distortions ranged between 0.1 and 0.2 mm for all organs at risk. Active shimming was estimated to reduce distortions by a factor of 2.3 to 2.9. A safety margin of 1 mm would have encompassed 99.8% of the distortions. Abstract: Background and purpose: Magnetic resonance imaging (MRI) is a crucial factor in optimal treatment planning for stereotactic radiosurgery. To further the awareness of possible errors in MRI, this work aimed to investigate the magnitude of susceptibility induced MRI distortions for intracranial organs at risk (OARs) and test the effectiveness of actively shimming these distortions. Materials and methods: Distortion maps for 45 exams of 42 patients (18 on a 1.5 T MRI scanner, 27 on a 3 T MRI scanner) were calculated based on a high-bandwidth double-echo gradient echo sequence. The investigated OARs were brainstem, chiasm, eyes, and optic nerves. The influence of active shimming was investigated by comparing unshimmed 1.5 T data with shimmed 3 T data and comparing the results to a model based prediction. Results: The median distortion for the different OARs was found to be between 0.13 and 0.18 mm for 1.5 T and between 0.11 and 0.13 mm for 3 T. The maximum distortion was found to be between 1.3 and 1.7 mm for 1.5 T and between 1.1 and 1.4 mm for 3 T. The variation of values was much higher for 1.5 T than for 3 T across all investigated OARs. Active shimming was found to reduce distortions by a factor of 2.3 to 2.9 compared to the expected values. Conclusions: Using a safety margin for OARs of 1 mm would have encompassed 99.8% of the distortions. Since distortions are inversely proportional to the readout bandwidth, they can be further reduced by increasing the bandwidth. Additional error sources like gradient nonlinearities need to be addressed separately.