Susceptibility-based positive contrast MR imaging exhibits excellent efficacy for visualizing the MR compatible metallic devices, by taking advantage of their high magnetic susceptibility ⁽¹⁾. However, data acquisition of the original technology is based on the spin echo(SE) sequence (SE-based technique) ⁽²⁾, resulting in a low sampling efficiency. To address this issue, a novel method is developed to accelerate the susceptibility-based positive contrast MR imaging.

Sequence and Image Reconstruction: A modified turbo spin echo (TSE) sequence is developed to accelerate data acquisition. Two datasets are acquired (with or without echo shift) for measuring the field induced by the device. TSE with shifted readouts (by a small amount, T_shift in the range of 0.2~0.7ms), is used to mitigate rapid signal dephasing and phase wrapping due to the highly susceptibility of the metallic devices. The amount of phase change induced by the local susceptibility difference between the metallic devices and tissues surround is accumulated during T_shift instead of the entire TE. After the data acquisition, the susceptibility mapping is calculated by using a kernel deconvolution algorithm with a regularized ℓ1 minimization to realize the positive contrast image, which is similar to the references ^(2,3).

Experiments: Two sets of experimental data were acquired on a 3T MRI scanner (SIEMENS Tim Trio, Germany). The first dataset was acquired using a titanium hollow biopsy needle (2.0 mm in diameter) which was inserted into a water phantom doped with 1.0 g/L copper sulfate solution. The needle was positioned perpendicular to B₀ and the axial multi-slice data were acquired with or without T_shift of 0.6 ms. Scan parameters are: TR= 2000 ms, TE = 18 ms, slice thickness = 1.5 mm, slice gap = 25%, in-plane resolution: 0.625×0.625 mm², and BW=134 Hz/pixel. The second phantom experiment was carried out by placing five brachytherapy seeds with different spacing into porcine tissue. A plastic, bamboo toothpick and an animal bone were respectively inserted to simulate the tissue cavity, capillary and human bone. The seeds orientation and imaging parameters were the same as those of the first experiment, except the slice thickness was set to 1.5 mm with no gap.

Results of the two experiments showed that the positive contrast imaging (Fig. 1c & Fig. 2c) of the proposed method were comparable to those of the SE-based technique (Fig. 1d & Fig. 2d) and CT (Fig. 1e & Fig. 2e), even though the CT could clearly show the cavity of the needle due to its much higher spatial resolution. The MIPs reconstructed obtained by the proposed method (Fig. 1f & Fig. 2f) demonstrated that good visualization of needle and seeds was realized. The brachytherapy seeds for the proposed method and SE-based technique were clearly identified from the bamboo toothstick (red), plastic stick (blue) and bone (green) (Fig. 2f&g), which were hard to differentiate them on the magnitude (Fig. 2b). Besides, the distances L1, L2 and L3 measured from the positive MR images were very close to the real values and the gold standard CT (Table 1).An accelerated susceptibility-based positive contrast imaging technique was developed and evaluated with phantom studies. Compared to the SE-based technique, the proposed method provides comparable positive contrast imaging results with significantly accelerating data acquisition of 3~4 times. Besides, similar to the SE-based technique, the proposed method can correctly differentiate the metallic devices from other structures and accurately visualize and localize the metallic devices, which were confirmed by CT.