Assessment of the Accuracy of Stereotactic Target Localization Using Magnetic Resonance Imaging: A Phantom Study

Abstract

Magnetic resonance imaging is a valuable technique for brain target visualization and localization for stereotactically guided open surgery, stereotactic biopsy, or stereotactic radiosurgery. However, questions remain regarding the extent of image distortion (caused mainly by inhomogeneity of the constant magnetic field induced by the imaged object and nonlinearities in the gradient fields) and its effect on the accuracy of stereotactic localization. A phantom study has been carried out to assess the accuracy of stereotactic localization using spin echo T1-weighted, T2-weighted, proton density, and gradient echo three-dimensional magnetic resonance imaging. A special cubical perspex phantom with the insert of an array of 81 solid perspex rods (2 mm in diameter and spaced 15 mm) was constructed and attached to the base of a Leksell sterotactic frame. The deviations between sterotactic coordinates based on magnetic resonance imaging determined in the treatment planning system and real geometrical position given by the construction of an array of perspex rods within the phantom were evaluated in a series of axial and coronal images for the above-mentioned four sequences. No dependence of the extent of deviations on the investigation sequences was observed. The image orientation and spatial position of measured points in the volume of cubical phantom were also not significantly influenced. The maximal deviation was observed for T1-weighted coronal study, 1.8 mm in the Z coordinate. However, average deviations in all of the eight performed studies were less than or equal to 0.6 mm. Phantom measurements proved minimal distortion effects for all investigated modalities and therefore no special corrections were applied for stereotactic localization.