Development of an anthropomorphic spine phantom suitable for fusion of MR Neurography with interventional flat-panel CT

Publication date: Available online 21 January 2019

Source: European Journal of Radiology

Author(s): Adrian Kobe, Matthias Zadory, Qeumars M. Hamie, Johannes M. Froehlich, Markus Klarhöfer, Thilo Elsässer, Thomas Pfammatter, Roman Guggenberger

Abstract

Purpose

To design a spine phantom suitable for fusion of MR neurography (MRN) with interventional flat panel computed tomography (FPCT) images from tissue-equivalent agarose gels and artificial nerves in MRI, including material with equal attenuation to bone in computed tomography (CT).

Methods

T1-/T2-relaxation times of target tissue were determined in vivo (n = 5) using MR mapping-techniques. Serial dilution of castor oil lipogels was performed ex vivo in order to define correct composition for tissue-equivalent relaxation times. Similarly, serial dilution series of calcium carbonate (CaCO₃) and barium sulphate (BaSO₄) in synthetic resin were used to adjust radiodensity of selected vertebral bodies (L1-L5) and sacrum in CT. Nerve tissue was simulated with agarose-impregnated polyethylene fibers. Spine phantom was assembled using respective components in anthropomorphic geometry. A fat-saturated, T2-weighted 3D SPACE STIR sequence was acquired for MRN and subsequently fused with an on-site FPCT scan of the phantom.

Results

In vivo T1-/T2-values for fat tissue were found to be at 394 ± 16 ms and 161 ± 16 ms, corresponding to a castor oil concentration of 50%. Analogously, bone marrow-equivalent values were measured at 822 ± 21 ms and 67 ± 6 ms, simulated with 40% castor oil. Cortical bone-like radiodensity of 1'115 ± 80 HU was achieved for artificial bone with 30% CaCO₃ and 1.5% BaSO₄. Simulated nerves were successfully depicted in MRN and fused with FPCT, combining optimal contrasts for nerves and bones on-site.

Conclusions

The customized phantom showed analogous tissue contrasts to in vivo conditions in both MRN and FPCT, facilitating simulations of fusion-image guided spine interventions.

from #Head and Neck by Sfakianakis via simeraentaxei on Inoreader http://bit.ly/2T77NMV