In vitro assessment of artifacts from commercially available markers for image-guided preoperative marking of bone and soft tissue lesions.

PURPOSE: To evaluate the magnetic resonance (MR) imaging artifacts from several commercially available MR imaging-compatible and non-MR imaging-compatible markers for image-guided marking of soft tissue lesions at different field strengths and sequences. MATERIALS AND METHODS: Ten different marking devices (anchor, coil, V, and X shapes) were injected into solid agar and imaged at two different field strengths (1.5 T, 3.0 T) and with three MR sequences (T1-weighted three-dimensional fast low-angle shot, T1-weighted two-dimensional fast low-angle shot MR fluoroscopy, and T2-weighted short inversion time inversion-recovery). Artifacts were assessed quantitatively by evaluating the areas of signal extinction and calculating equivalent diameters and qualitatively by visual grading. Paired t tests were used to assess the influence of sequence types and field strengths. RESULTS: Artifact diameters ranged from negligible to 45.8 mm. Sequence types and field strength did not have a statistically relevant influence on the artifact size. All artifacts from non-MR imaging-compatible V- and X-shaped markers were rated as "too large;" the artifacts from MR imaging-compatible V- and X-shaped markers and from all anchor-shaped devices were rated as "too small." The highest ratings were achieved by an MR imaging-compatible coil-shaped marker. CONCLUSIONS: The artifacts of the tested marking devices are mainly determined by their shapes and materials. Sequence types and field strengths have only a small influence on the artifacts. The most distinguished signal voids with an intermediate size resulted from an MR imaging-compatible coil-shaped marker.

Biopsy needle tips with markers--MR compatible needles for high-precision needle tip positioning.

Needle tip visualization is of high importance in magnetic resonance imaging (MRI) guided interventional procedures, for example for taking biopsies from suspicious lesions in the liver or kidney. The exact position of the needle tip is often obscured by image artifacts arising from the magnetic properties of the needle. The authors investigated two special biopsy needle tip designs using diamagnetic coatings. For common interventional MR sequences, the needle tip can be identified in the MR image by several equidistant dark spots arranged along a straight line. A dotted instead of a solid line allows for an improved control of the movement of the needle, not only if the needle is tilted toward the imaging plane, but also if the needle leaves an empty canal with signal extinction, which cannot be distinguished from the needle material itself. With the proposed design the position of the needle tip can be estimated with a precision of approximately 1 mm using conventional FLASH, FISP, and TSE sequences, as used for interventional MR. Furthermore, the size of the biopsy probe can be estimated from the artifact. In using needles with a properly designed tip coating, taking biopsies under MR control is beginning to be greatly simplified. The approach to design artifacts using diamagnetic material in combination with paramagnetic material paves the way toward new instruments and implants, suitably tailored to the needs of the interventional radiologist.

Numerical simulations of intra-voxel dephasing effects and signal voids in gradient echo MR imaging using different sub-grid sizes.

Signal void artifacts in gradient echo imaging are caused by the intra-voxel dephasing of the spins. Intra-voxel dephasing can be estimated by computing the field distribution on a sub-grid inside each picture element, followed by integration of all magnetization components. The strategy of computing the artifacts based on the integration of the sub-voxel signal components is presented here for different sub-grids. The coarseness of the sub-grid is directly related to computational effort. The possibility to save memory space and computing time for the dipole model by computing the field only on a sub-grid is addressed in the presented article. It is investigated as to how far computational time and memory space can be reduced by using an appropriate sub-grid. Numerical results for a model of a partially diamagnetically coated needle shaft are compared to experimental findings. In the case of a pure titanium needle, it is shown as being sufficient to compute the field distribution on a sub-grid that is at least four times coarser in each direction than the grid used to discretize the object in the related MR image. Due to three nested loops over the 3D grid, the need for memory space and time is saved by a factor 64. Deviations between measurements and simulations for the broad side of the artifact (uncompensated) and for the small side of the artifact (compensated) were 15.5%, respectively, 19.1% for orientation parallel to the exterior field, and 22.7%, respectively, 23.1% for orientation perpendicular to the exterior field.