Metal implant imaging using highly undersampled phase-cycled 3D bSSFP
Damien Nguyen1,2, Tom Hilbert3,4,5, Jean-Philippe Thiran5,6, Tobias Kober3,4,5, and Oliver Bieri1,2

1Radiological Physics, Dep. of Radiology, University of Basel Hospital, Basel, Switzerland, 2Department of Biomedical Engineering, University of Basel, Basel, Switzerland, 3Advanced Clinical Imaging Technology (HC CMEA SUI DI BM PI), Siemens Healthcare AG, Lausanne, Switzerland, 4Department of Radiology, University Hospital (CHUV), Lausanne, Switzerland, 5LTS5, École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland, 6Department of Radiology, University Hospital Lausanne (CHUV), Lausanne, Switzerland


In this study, we explore the possibility of using a highly undersampled 3D phase-cycled balanced Steady-State Free-Precession (bSSFP) sequence (trueCISS) to image metal implants in the body and compare it to the Slice Encoding for Metal Artifact Correction (SEMAC) method. We show that the trueCISS approach not only offers qualitatively good morphological images, but also delivers quantitative maps that could potentially improve the overall diagnostic quality and efficiency within a clinically reasonable time.


It is important to monitor potential pathological tissue changes after implanting a metal prosthesis, e.g. a hip implant. The current gold standard to perform such an exam with MRI is “Slice Encoding for Metal Artifact Correction” (SEMAC) [1], a method enabling the acquisition of images with high robustness against metal-induced artifacts due to strong magnetic field inhomogeneity. In this work, we evaluate the use of another recently proposed highly undersampled phase-cycled 3D balanced Steady-State Free Precession sequence dubbed trueCISS [2] in order to obtain higher resolution images and improved contrast within the same overall acquisition time. trueCISS is tested in vitro against the gold standard technique SEMAC.

Materials & Methods

A titanium hip prosthesis embedded in an agar gel with the addition of 0.2 mM of ferumoxide was acquired at 1.5T (MAGNETOM Avanto, Siemens Healthcare, Germany) using an optimized 2D multi-slice SEMAC [1] sequence (0.9x0.9x3mm3, TR 700 ms, TE 4.9 ms, flip-angle 130°, TA 4:51) and a prototype trueCISS sequence [2] (0.9x0.9x0.9mm3, TR 3.59 ms, TE 1.80 ms, flip-angle 15°, 16 phases $$$\phi$$$ = 0°, 22.5°, 45°, …, 337.5°, 8-fold undersampling, TA 4:47). First, the phase-cycled images were reconstructed using a sparse iterative reconstruction. Subsequently, the bSSFP signal-model (cf. equation below with $$$M_0$$$ equilibrium magnetization, $$$\Lambda$$$ relaxation time ratio $$$T_1/T_2$$$, $$$\alpha$$$ flip-angle, $$$\Delta\phi$$$ local phase offset relating to field inhomogeneity) was fitted voxel-wise onto these images, effectively estimating three parameter maps: $$$M_0$$$, $$$\Lambda$$$, and $$$\Delta\phi$$$. The resulting parameter maps were used to synthesize an on-resonant bSSFP signal image by applying the forward signal model with $$$\phi - \Delta\phi$$$ = 0°.

$$M = M_0 \left\lvert\frac{2\sin\alpha \cos\left(\frac{\phi - \Delta\phi}{2}\right)}{1 + \cos\alpha + 2\cos\left(\phi - \Delta\phi\right) + \left(4\Lambda - \cos^2\left(\phi-\Delta\phi\right)\sin^2\left(\frac{\alpha}{2}\right)\right)}\right\rvert$$

Results and discussion

Fig. 1a shows the result of a SEMAC acquisition of the titanium metal implant in gel. The image was interpolated in the Z-direction to match the trueCISS resolution. Residual artefacts due to off-resonances can be seen around the cap as well as around the shoulder and tip of the metal implant (arrows). The images of a trueCISS acquisition on the same phantom are presented in Fig 1b, showing a slice at the approximately same location as the SEMAC image. The trueCISS approach offers similar image quality with the same overall scan time, although it has a three times higher resolution along the slice direction. While most of the banding artefacts have been eliminated in the trueCISS image, some remaining artifacts are visible at the cap, the neck as well at the tip of the prosthesis (arrows). In addition, the trueCISS image retains some information about changes in gel density which are invisible on the SEMAC image due to the choice of imaging parameters.

Fig. 2 presents the parameter maps for $$$M_0$$$, $$$\Lambda$$$, and $$$\Delta\phi$$$ estimated by trueCISS. The $$$\Lambda$$$ map is of particular interest as we expect it to provide good contrast between muscle tissues ($$$\Lambda\approx 10-20$$$), fat ($$$\Lambda \approx 3$$$) and water ($$$\Lambda\approx 1$$$). This could be relevant for assessing the presence of fat, lesions or edemas in regions near the implant.

A limitation of the present approach is that we rely on measuring the true bSSFP signal in each voxel, a condition not necessarily satisfied in voxels very close to the implant. In those regions, it is likely that very strong local susceptibility gradients within the voxels induce a constant dephasing of the spins during the TR, which may lead to a transition of the steady-state signal from balanced to unbalanced or even spoiled SSFP. However, these effects can be mitigated by choosing smaller voxel sizes as was done in the present study.


trueCISS imaging is able to provide nearly artifact-free bSSFP images with similar overall acquisition time and quality as the SEMAC sequence. The main advantage of the present approach over other conventional methods is that it not only delivers clinically relevant morphological images in a time-efficient manner, but also provides quantitative information that can help improve diagnostic quality and accuracy.


No acknowledgement found.


1. W. Lu, K.B. Pauly, G.E. Gold, J.M. Pauly and B.A. Hargreaves. SEMAC: Slice Encoding for Metal Artifact Correction in MRI. Magn Reson Med 2009;62(1):66–76

2. T. Hilbert, D. Nguyen, T. Kober, J.-P.Thiran, G. Krueger and O. Bieri. TrueCISS: Genuine bSSFP Signal Reconstruction from Undersampled Multiple-Acquisition SSFP Using Model-Based Iterative Non-Linear Inversion. Proc. Intl. Soc. Mag. Reson. Med. Toronto, Canada 2015


Fig. 1 SEMAC and reconstructed on-resonant trueCISS image of a titanium prosthesis embedded in an agar gel.

Fig. 2 trueCISS estimated parameter maps ($$$M_0$$$, $$$\Lambda$$$, and $$$\Delta\phi$$$ resp.) from phase-cycled bSSFP acquisitions.

Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)