Functional Laterality of Task-Evoked Activation in Sensorimotor Cortex of Preterm Infants: An Optimized 3 T fMRI Study Employing a Customized Neonatal Head Coil.

BACKGROUND: Functional magnetic resonance imaging (fMRI) in neonates has been introduced as a non-invasive method for studying sensorimotor processing in the developing brain. However, previous neonatal studies have delivered conflicting results regarding localization, lateralization, and directionality of blood oxygenation level dependent (BOLD) responses in sensorimotor cortex (SMC). Amongst the confounding factors in interpreting neonatal fMRI studies include the use of standard adult MR-coils providing insufficient signal to noise, and liberal statistical thresholds, compromising clinical interpretation at the single subject level. PATIENTS / METHODS: Here, we employed a custom-designed neonatal MR-coil adapted and optimized to the head size of a newborn in order to improve robustness, reliability and validity of neonatal sensorimotor fMRI. Thirteen preterm infants with a median gestational age of 26 weeks were scanned at term-corrected age using a prototype 8-channel neonatal head coil at 3T (Achieva, Philips, Best, NL). Sensorimotor stimulation was elicited by passive extension/flexion of the elbow at 1 Hz in a block design. Analysis of temporal signal to noise ratio (tSNR) was performed on the whole brain and the SMC, and was compared to data acquired with an 'adult' 8 channel head coil published previously. Task-evoked activation was determined by single-subject SPM8 analyses, thresholded at p < 0.05, whole-brain FWE-corrected. RESULTS: Using a custom-designed neonatal MR-coil, we found significant positive BOLD responses in contralateral SMC after unilateral passive sensorimotor stimulation in all neonates (analyses restricted to artifact-free data sets = 8/13). Improved imaging characteristics of the neonatal MR-coil were evidenced by additional phantom and in vivo tSNR measurements: phantom studies revealed a 240% global increase in tSNR; in vivo studies revealed a 73% global and a 55% local (SMC) increase in tSNR, as compared to the 'adult' MR-coil. CONCLUSIONS: Our findings strengthen the importance of using optimized coil settings for neonatal fMRI, yielding robust and reproducible SMC activation at the single subject level. We conclude that functional lateralization of SMC activation, as found in children and adults, is already present in the newborn period.

MR imaging of healthy knees in varying degrees of flexion using a stretchable coil array provides comparable image quality compared to a standard knee coil array.

OBJECTIVE: Stretchable coils allow knee imaging at varying degrees of flexion. Purpose was to compare a new-developed stretchable 8-channel to a standard 8-channel knee coil array by means of quantitative and qualitative image analysis. MATERIAL AND METHODS: IRB approved prospective study. Knee MR imaging in 10 healthy volunteers was performed at 3T using a standard 8-channel and a new-developed stretchable 8-channel coil array at 0°, 45°, and 60° of flexion and at 0° (standard coil). Image parameters were identical. Signal-to-noise ratio (SNR) was determined by combining the images with separately acquired noise data on a pixel-by-pixel basis using MATLAB routines (Natick, MA, USA). Images were qualitatively analysed by two independent radiologists who graded the visibility of several anatomic structures from 1=not visible to 5=excellent. ANOVA, Wilcoxon and kappa statistics were used. RESULTS: Mean SNR±standarddeviation of bone was 54.7±10.4 and of muscle 28.0± 4.4 using the stretchable coil array and 54.6±8.2 and 33.4±4.5, respectively, using the standard knee coil array. No statistically significant SNR differences were found between both arrays (bone, p=0.960; muscle, p=0.132). SNR was not degraded at higher degrees of flexion. The qualitative image analysis did not reveal statistically significant differences between the stretchable and standard coil array with regard to the visibility of anatomic structures (p=0.026-1.000). Overall kappa was 0.714. CONCLUSION: Stretchable 8-channel coil arrays provide similar SNR and visibility of anatomic structures compared to standard 8-channel knee coil arrays. MR imaging with high SNR will now be possible in flexed knees.

Do dynamic-based MR knee kinematics methods produce the same results as static methods?

MR-based methods provide low risk, noninvasive assessment of joint kinematics; however, these methods often use static positions or require many identical cycles of movement. The study objective was to compare the 3D kinematic results approximated from a series of sequential static poses of the knee with the 3D kinematic results obtained from continuous dynamic movement of the knee. To accomplish this objective, we compared kinematic data from a validated static MR method to a fast static MR method, and compared kinematic data from both static methods to a newly developed dynamic MR method. Ten normal volunteers were imaged using the three kinematic methods (dynamic, static standard, and static fast). Results showed that the two sets of static results were in agreement, indicating that the sequences (standard and fast) may be used interchangeably. Dynamic kinematic results were significantly different from both static results in eight of 11 kinematic parameters: patellar flexion, patellar tilt, patellar proximal translation, patellar lateral translation, patellar anterior translation, tibial abduction, tibial internal rotation, and tibial anterior translation. Three-dimensional MR kinematics measured from dynamic knee motion are often different from those measured in a static knee at several positions, indicating that dynamic-based kinematics provides information that is not obtainable from static scans.

In vitro and in vivo comparison of wrist MR imaging at 3.0 and 7.0 tesla using a gradient echo sequence and identical eight-channel coil array designs.

PURPOSE: To evaluate in vivo MR imaging of the wrist at 3.0 Tesla (T) and 7.0T quantitatively and qualitatively. MATERIALS AND METHODS: To enable unbiased signal-to-noise ratio (SNR) comparisons, geometrically identical eight-channel receiver arrays were used at both field strengths. First, in vitro images of a phantom bottle were acquired at 3.0T and 7.0T to obtain an estimate of the maximum SNR gain that can be expected. MR images of the dominant wrist of 10 healthy volunteers were acquired at both field strengths. All measurements were done using the same sequence parameters. Quantitative SNR maps were calculated on a pixel-by-pixel basis and analyzed in several regions-of-interest. Furthermore, the images were qualitatively evaluated by two independent radiologists. RESULTS: The quantitative analysis showed SNR increases of up to 100% at 7.0T compared with 3.0T, with considerable variation between different anatomical structures. The qualitative analysis revealed no significant difference in the visualization of anatomical structures comparing 3.0T and 7.0T MR images (P>0.05). CONCLUSION: The presented results establish the SNR benefits of the transition from 3.0T to 7.0T for wrist imaging without bias by different array designs and based on exact, algebraic SNR quantification. The observed SNR increase nearly reaches expected values but varies greatly between different tissues. It does not necessarily improve the visibility of anatomic structures but adds valuable latitude for sequence optimization.

NMR probes for measuring magnetic fields and field dynamics in MR systems.

High-resolution magnetic field probes based on pulsed liquid-state NMR are presented. Static field measurements with an error of 10 nanotesla or less at 3 tesla are readily obtained in 100 ms. The further ability to measure dynamic magnetic fields results from using small ( approximately 1 microL) droplets of MR-active liquid surrounded by susceptibility-matched materials. The consequent high field homogeneity allows free induction decay signals lasting 100 ms or more to be readily achieved. The small droplet dimensions allow the magnetic field to be measured even in the presence of large gradients. Highly sensitive detection yields sufficient SNR to follow the relevant field evolution without signal averaging and at bandwidths up to hundreds of kHz. Transient, nonreproducible effects and drifts are thus readily monitored. The typical application of k-space trajectory mapping has been demonstrated. Potential further applications include characterization, tuning, and maintenance of gradient systems as well as the mapping of the static field distribution of MRI magnets. Connection of the probes to a standard MR spectrometer is similar to that used for imaging coils.

Modular design of receiver coil arrays.

We describe a modular and hence flexible system for connecting MR surface coils to create a receiver array. Up to 16 individual coils of different size and shape depending on the application are plugged into a connector box that houses the control electronics. Preamplification, matching and detuning circuitry are housed on a circuit board directly attached to each coil loop. Electrical adjustments for tuning or decoupling for each coil configuration are not needed thanks to effective preamplifier decoupling provided through a Pi matching network. Radio-frequency safety and electrically stable cabling are ensured by multiple radio-frequency traps. Array modules for 1.5 and 3 T have been simulated, constructed, tested, and used for imaging experiments.