Implementation of a 0.55T MR simulator for intracranial stereotactic radiosurgery

Authors: Kim, Joshua; Rusnac, Robert; Dicarlo, Amanda; Doemer, Anthony; Ruiz, Ralph; Siddiqui, M Salim; Shah, M; Mellon, Eric A; Movsas, Benjamin; Thind, Kundan

Source journal: Journal of applied clinical medical physics (2026-12-01)

DOI: 10.1002/acm2.70909

Abstract

BACKGROUND: Low-field MRI has several advantages for radiation therapy simulation. The reduced field strength lowers susceptibility artifacts near metallic implants, decreases intrinsic geometric distortion, and improves safety through lower SAR, reduced projectile risk, and fewer device interactions. MRI is the imaging foundation for SRS target delineation, and initial low-field MRI feasibility for brain radiotherapy planning has been demonstrated in small cohorts. However, SRS clinical implementation has not been reported. We describe the implementation of a 0.55T MR simulator and characterize lesion conspicuity across 102 lesions in 13 patients with multiple brain metastases, drawn from 211 consecutive SRS patients treated to 511 lesions over 12 months. PURPOSE: To describe the clinical implementation of a 0.55T MR simulator (MAGNETOM Free.Max RT Edition, Siemens Healthineers) as a dedicated radiation therapy simulation system, following a systematic framework of acceptance testing per AAPM TG-284, longitudinal quality assurance, and SRS MRI protocol optimization. METHODS: Acceptance testing and commissioning followed AAPM TG-284 recommendations. Testing included the use of vendor-provided homogeneous phantoms, the ACR large phantom, and a commercially available large-field geometric distortion phantom. Longitudinal quality assurance was performed over a 12-month period using the ACR large phantom. Measured parameters included geometric accuracy, percent image uniformity (PIU), percent signal ghosting, low contrast detectability (LCD), and SNR. Contrast to noise ratio (CNR) was calculated for the lesions of a subset of SRS patients having 4 or greater metastatic lesions to evaluate the conspicuity of small SRS lesions in this challenging environment. RESULTS: All acceptance testing met TG-284 recommendations. Central frequency and transmitter gain were stable with changes of less than 0.01% and 0.3%, respectively, over the first month of testing. Geometric distortion after vendor 3D corrections measured less than 1 mm within 10 cm and less than 2 mm within 20 cm of isocenter. Monthly phantom testing over 12 months confirmed geometric stability within established tolerances. A brain imaging protocol achieved 1 mm slice thickness on 3D T1-weighted acquisitions consistent with ISRS guidelines, with total scan time under 35 minutes. To date, 211 intracranial SRS patients have been scanned and treated to a total of 511 lesions with a median volume of 0.58cc [0.004 - 304.48 cc]. The median CNR for the 102 lesions in the set of 13 patients with 4 or greater metastatic lesions was calculated as 6.81 [1.7-16.9] with only two lesions having a CNR < 2. CONCLUSIONS: These findings support the feasibility of 0.55T MRI as a dedicated RT simulator capable of meeting the imaging demands of intracranial SRS. Formal comparative studies against higher field systems are needed before low-field imaging can be considered a standalone replacement for SRS planning.

Read the full text

Related topics

Related open access articles

Abstract and bibliographic data from PubMed (U.S. National Library of Medicine). Copyright of the abstract belongs to the original publisher or authors.