Phantom-based characterization of distortion on a magnetic resonance imaging simulator for radiation oncology

Abstract

One of the major issues potentially limiting treatment planning with solely MR images is the possibility of geometric distortion inherent in MR images. We designed a large distortion phantom containing a 3D array of spheres and proposed a three-dimensional (3D) approach to determine the distortion of MR image volume. The approach to overcome partially filled spheres is also presented. The phantom was assembled with a 3D array of spheres filled with contrast and was scanned with a 3T MRI simulator. A 3D whole-sphere or half-sphere template is used to match the image pattern. The half-sphere template is used when the normalized cross-correlation value for the whole-sphere template is below a predetermined threshold. Procrustes method was applied to remove the shift induced by rotation and translation of the phantom. Then the distortion map was generated. Accuracy of the method was verified using CT images of a small phantom of the same design. The analysis of the small phantom showed that the method is accurate with an average offset of estimated sphere center 0.12 ± 0.04 mm. The Procrustes analysis estimated the rotation angle to be 1.95° and 0.01°, respectively, when the phantom was placed at 2° and 0° from the ceiling laser. The analysis showed that on the central plane through the magnet center, the average displacement is less than 1 mm for all radii. At distal planes, when the radius is less than 18 cm, the average displacement is less than 1 mm. However, the average displacement is over 1 mm but still less than 1.5 mm for larger radii. A large distortion phantom was assembled and analysis software was developed to characterize distortions in MRI scans. The use of two templates helps reduce the potential impact of residual air bubbles in some of the spheres.

Characterization of spatial distortion in a 0.35 T MRI-guided radiotherapy system

“…The spheres and connecting rods were carved out of the acrylic plates using a CNC machine, with a nominal precision of 0.025 mm in each of the three dimensions (Huang et al 2016). The matrix was filled with 0.4 mM NiCl aqueous solution that results in a field inhomogeneity of 0.85 ppm due to its molar magnetic susceptibility of 6145 × 10 −6 cm 3 mol −1 (Huang et al 2016). The presence of air bubbles in the phantom was noted by Huang et al, potentially reducing the accuracy of sphere localization.…”

“…In order to minimize bubbles, the matrix was filled by joining and sealing the plates in an acrylic superstructure while entirely submerged in NiCl solution. Additionally, the manufacturer increased the diameter of the spheres from 7 mm (as in Huang et al (2016)) to the present value of 8 mm. The sphere located in the center row, slice and column was aligned to isocenter.…”

Image-based gradient non-linearity characterization to determine higher-order spherical harmonic coefficients for improved spatial position accuracy in magnetic resonance imaging

“…In each of these applications, the soft tissue contrast of MRI is desired in order to obtain the necessary information, but the spatial distortion of the MR image presents a barrier to adoption[6,7]. …”

Gradient nonlinearity calibration and correction for a compact, asymmetric magnetic resonance imaging gradient system

“…If linear gradients are presumed during image reconstruction, the effects of gradient nonlinearity (GNL) will manifest as geometric distortion into the generated images (O’Donnell and Edelstein 1985; Glover and Pelc 1986; Schad et al 1992; Janke et al 2004; Doran et al 2005; Baldwin et al 2007; Baldwin et al 2009). The GNL-induced distortion has substantial impact on applications demanding high geometric accuracy, such as radiation therapy planning (Chen et al 2006; Huang et al 2016), apparent diffusion coefficient mapping (Tan et al 2013), and longitudinal studies of neurodegenerative diseases (Han et al 2006; Jovicich et al 2006; Gunter et al 2009). If the GNL fields are a priori known, their effects may be retrospectively corrected in image domain after MRI reconstruction (Glover and Pelc 1986), as conventionally implemented on commercial MR systems.…”

“…Therefore, only model coefficients of odd-orders are non-negligible and required for distortion correction. Alternatively, phantom-based calibration methods have also been described to extract GNL information from MR images acquired on fiducial phantoms based on various mathematical models, such as spherical harmonic polynomial model, spline model, and polynomial model (Wang et al 2004; Doran et al 2005; Hwang et al 2012a; Hwang et al 2012b; Trzasko et al 2015; Huang et al 2016). Such methods allow the calibration of system specific GNL distortion fields.…”