A pneumatic phantom for mimicking respiration-induced artifacts in spinal MRI.

PURPOSE: To design a phantom capable of mimicking human respiration to serve as a testing platform for correction of the static and time-evolving magnetic field distortions typically encountered in MRI of the spinal cord. METHODS: An inflation system to mimic the air variation of the human lungs was constructed. The inflation system was linked to a phantom containing synthetic lungs and an ex vivo human spine. The relationship between air pressure and phantom lung volume was evaluated via imaging experiment. The geometric distortion (pseudo-displacement) caused by the B0 inhomogeneities was measured on echo planar imaging slices for different air volumes. RESULTS: Linear and quadratic relations linking air pressure to phantom lung volume were observed with a Pearson correlation coefficient of 0.99. Air distribution was uneven across the synthetic lungs, exhibiting a left-to-right lung volume ratio of up to 5/4. The pseudo-displacement artifact of the spine caused by the air-filled lungs was observed. CONCLUSION: The proposed phantom can reproduce the lung volume variation of human respiration and thus can serve as a reliable testing platform for the correction of the associated time-varying B0 field distortions. Details of the construction and code for the inflation system microcontroller are available for download as open source. Magn Reson Med 79:600-605, 2017. © 2017 International Society for Magnetic Resonance in Medicine.

A 24-channel shim array for the human spinal cord: Design, evaluation, and application.

PURPOSE: A novel multichannel shim array is introduced to improve MRI and spectroscopic studies of the human spinal cord. METHODS: Twenty-four-channel shim and 8-channel transceiver arrays were designed to insert into the patient bed table to lie in close proximity to the subject's spine. The reference field patterns of each of the shim channels (Hz/A) were determined empirically via gradient echo field mapping and subsequently used to demonstrate shim performance at 3 Tesla using an ex vivo phantom, which incorporated a fixed human spine. The shim was further demonstrated on five healthy volunteers. RESULTS: Application of the shim to the ex vivo phantom reduced the standard deviation of the field over the spinal volume of interest (123.4 cm3 ) from an original 51.3 Hz down to 32.5 Hz, amounting to an improvement in field homogeneity of 36.6%. In vivo, the spine shim resulted in an average improvement in field homogeneity of 63.8 ± 15.4%. CONCLUSION: The localized spine shim offers a promising new means of correcting magnetic field distortion in the spinal cord. Magn Reson Med 76:1604-1611, 2016. © 2016 International Society for Magnetic Resonance in Medicine.

SHARP edges: recovering cortical phase contrast through harmonic extension.

PURPOSE: To recover local phase contrast at the edges of the brain (e.g., cortex), where it is otherwise unavailable with the conventional form of the technique sophisticated harmonic artifact reduction for phase data (SHARP). METHODS: A harmonic potential field, such as the magnetic "background" field, is an analytic field and can thus be represented by a convergent power series. Using SHARP to obtain an initial estimate of the harmonic background field over a reduced inner portion of the brain, a three-dimensional Taylor expansion was performed to extend field coverage to the brain edges. The method, called Extended-SHARP, was quantitatively assessed through a numerical field-forward simulation and qualitatively demonstrated in vivo. RESULTS: Using a typical spherical kernel (6 mm radius), Extended-SHARP recovered on average 26% more in vivo brain volume than SHARP. When applied to the numerical model, local field contrast around an otherwise lost edge source was recovered, with the resulting error comparable to that of conventional SHARP. CONCLUSION: The lost field values near the edges of the brain can be recovered through an easily implemented adaptation to conventional SHARP.