Evaluation of Geometric Distortion In MRI-Only Interstitial Brachytherapy Treatment Plans at 3 Tesla
Abstract
Purpose
Geometric distortions in MRI increase with distance from the magnetic isocenter and may impact spatial accuracy in gynecological interstitial brachytherapy (ISBT). This is particularly relevant for ISBT techniques, such as the Syed-Neblett template, where needles frequently extend laterally to treat disease with significant lateral extent. This study quantifies MRI geometric distortion as a function of distance from isocenter and assesses implications for applicator reconstruction accuracy.
Methods
An interstitial gel phantom embedded with reference ball bearings (BBs) at known locations was imaged using MRI (T1-weighted sequence) and CT, with CT as the geometric ground truth. Point-to-reference distances between interstitial needles and BBs were measured on both imaging modalities and compared to quantify distortion on MRI. Distortion magnitude was evaluated as a function of radial distance from isocenter, characterizing spatial accuracy across the phantom and trends associated with increasing radial extent.
Results
T1-weighted MRI geometric distortion increased with radial distance from isocentre across the interstitial phantom, with substantial needle-specific variability. Distortion ranged from sub-millimeter near the isocentre to approximately 1-1.7 mm at radial distances up to 60 mm. While several needles exhibited a positive dependence on distance, others showed minimal distortion, reducing the overall average slope (~0.005 mm per mm). Greater variability was observed at larger distances.
Conclusion
MRI geometric distortion introduces measurable spatial inaccuracies in ISBT phantoms, increasing at greater radial distances. For ISBT applications requiring lateral needle placement, these distortions represent a non-negligible source of reconstruction uncertainty. Quantifying distortion as a function of radial extent provides guidance for evaluating MRI-only workflows and highlights the importance of distortion-aware imaging and reconstruction strategies in gynecological ISBT. Future work will evaluate the dosimetric impact of these spatial inaccuracies to determine clinical significance.