Investigation of Parameters for Metallic Artifact Reduction using 3-D SPACE in MR Lumbar Spine at 1.5 Tesla: A Phantom Study

Abstract

Metallic artifact reduction in magnetic resonance imaging (MRI) has increasingly interested especially metal implant postoperative in lumbar spine. Recently, the new pulse sequence based on true isotropic 3D voxel with high resolution namely Sampling Perfection with Application Optimized Contrasts using Different Flip Angle Evolutions (SPACE) was introduced. This study aimed to determine the appropriate 3D T2-SPACE parameters for reducing the metallic artifact in MR lumbar spine 1.5 Tesla. Cylindrical phantom filled with 3.3 g/L, NiCl2-6H2O and 2.4 g/L NaCl solution, and two commonly used metal implants (stainless steel and titanium) for lumbar spine were acquired using turbo spin echo (TSE) and SPACE pulse sequences in T2-weighted image contrast. A receiver bandwidth of 296, 501, 723 Hz/pixel and flip angle at 100°,150°, 180° were adjusted in order to determine the 3D T2-SPACE appropriate parameters. The volume of metal artifact was quantitatively assessed by applying a histogram distribution for obtaining the normal background signal intensity (mean value±3SD) of the solution. Such normal background range was applied to the metal artifact image dataset to determine the thresholding level. All signal values outside this normal range were calculated as susceptibility artifacts from the metal implants. The quantitative image quality was evaluated in terms of signal-to-noise ratio (SNR). The appropriate protocol was selected according to ranking score (1-10 points) by sorting the volume of metallic artifact, SNR and specific absorption rate (SAR).

By acquiring the phantom with the 3D T2 SPACE (protocol 1-9) and 2D T2 TSE (protocol 10) at 1.5 T, the results showed that the average metallic artifact volume of protocol 1-10 were 162.64±1.88, 158.30±1.79, 157.82±2.36, 126.43±1.22, 124.87±0.29, 122.93±1.57, 117.55±1.03, 114.82±1.55, 112.40±0.43 and 184.89±1.19 cm3 for the stainless-steel and for titanium, the average metallic artifact volume of protocol 1-10 were 33.80±1.03, 32.56±0.48, 33.79±0.90, 13.39±0.67, 13.77±0.73, 12.74±0.48, 10.89±0.77, 11.42±0.64, 10.67±0.71 and 38.86±0.97 cm3. The metal artifacts were highest reduced at protocol 9 by 39.21% for stainless steel and 72.55% for titanium.

In conclusions, using the 3D SPACE pulse sequence, the artifact volume can be effectively reduced by increasing the receiver bandwidth. The appropriate parameters for metallic artifact reduction with highest-ranking scores were protocol 4 and 5 with bandwidth 501 Hz/pixel, flip angle 100° and 150° in both of stainless steel and titanium respectively.