Acceleration of Magnetic Resonance Fingerprinting Reconstruction Using Denoising and Self-Attention Pyramidal Convolutional Neural Network

Magnetic resonance fingerprinting (MRF) based on echo-planar imaging (EPI) enables whole-brain imaging to rapidly obtain T1 and T2* relaxation time maps. Reconstructing parametric maps from the MRF scanned baselines by the inner-product method is computationally expensive. We aimed to accelerate the reconstruction of parametric maps for MRF-EPI by using a deep learning model. The proposed approach uses a two-stage model that first eliminates noise and then regresses the parametric maps. Parametric maps obtained by dictionary matching were used as a reference and compared with the prediction results of the two-stage model. MRF-EPI scans were collected from 32 subjects. The signal-to-noise ratio increased significantly after the noise removal by the denoising model. For prediction with scans in the testing dataset, the mean absolute percentage errors between the standard and the final two-stage model were 3.1%, 3.2%, and 1.9% for T1, and 2.6%, 2.3%, and 2.8% for T2* in gray matter, white matter, and lesion locations, respectively. Our proposed two-stage deep learning model can effectively remove noise and accurately reconstruct MRF-EPI parametric maps, increasing the speed of reconstruction and reducing the storage space required by dictionaries.

Vertical Field MR Imaging of Upper Thorax and Spine in Small Children

To improve image quality in a vertical field MR imaging unit, operating at low field strength (0.3 T), we have designed a half-elliptical coil for use in the upper thoracic region of small children. Our intention was also to shorten the examination time, which until now has been long, because several scans with different coils have been necessary to cover the thoracic region. The experimental coil is designed so that a child's shoulders fit into the central region. The coil consists of 2 serially connected cable-loops, mounted on a foam rubber vest. The coil performance was tested in a phantom and improvements relative to standard coils were demonstrated in in vivo studies. The results indicate that by using the half-elliptical coil, the signal-to-noise (S/N) ratio can be improved by a factor of 2 to 3 in the thoracic region of a child.

Comparison of Myelography, CT Myelography and Magnetic Resonance Imaging in Cervical Spondylosis and Disk Herniation

Twenty-six patients with cervical radiculopathy and/or myelopathy caused by spondylosis or disk herniation were examined with myelography, CT myelography and MR. Fourteen of the patients were operated upon and 11 of them underwent postoperative MR and CT. The three radiologic methods provided comparable information about narrowing of the subarachnoid space and compression of the spinal cord. It was more difficult to distinguish bone from soft tissue with MR only, but the combination of MR and conventional radiography gave sufficient information for this differentiation. When radiologic nerve root sheath deformity was compared with clinical radiculopathy, myelography, CT myelography and MR had similar sensitivity and accuracy. Postoperative MR could reveal remaining indentation on the thecal sac and the cord but CT without contrast medium was useful as a complement to determine the aetiology of the indentation. Because MR has several practical advantages, it is well suited as the primary imaging modality, together with conventional radiography, for the preoperative radiologic evaluation of patients with cervical radiculopathy and/or myelopathy. Postoperative MR is useful in patients with persistent or new symptoms.

Imaging of Herniated Discs of the Cervical Spine: Inter-Modality Differences between 64-Slice Multidetector CT and 1.5-T MRI

Objective

To assess inter-modality variability when evaluating cervical intervertebral disc herniation using 64-slice multidetector-row computed tomography (MDCT) and magnetic resonance imaging (MRI).

Materials and Methods

Three musculoskeletal radiologists independently reviewed cervical spine 1.5-T MRI and 64-slice MDCT data on C2-3 though C6-7 of 51 patients in the context of intervertebral disc herniation. Interobserver and inter-modality agreements were expressed as unweighted kappa values. Weighted kappa statistics were used to assess the extents of agreement in terms of the number of involved segments (NIS) in disc herniation and epicenter measurements collected using MDCT and MRI.

Results

The interobserver agreement rates upon evaluation of disc morphology by the three radiologists were in fair to moderate agreement (k = 0.39-0.53 for MDCT images; k = 0.45-0.56 for MRIs). When the disc morphology was categorized into two and four grades, the inter-modality agreement rates were moderate (k-value, 0.59) and substantial (k-value, 0.66), respectively. The inter-modality agreements for evaluations of the NIS (k-value, 0.78) and the epicenter (k-value, 0.79) were substantial. Also, the interobserver agreements for the NIS (CT; k-value, 0.85 and MRI; k-value, 0.88) and epicenter (CT; k-value, 0.74 and MRI; k-value, 0.70) evaluations by two readers were substantial. MDCT tended to underestimate the extent of herniated disc lesions compared with MRI.

Conclusion

Multidetector-row computed tomography and MRI showed a moderate-to-substantial degree of inter-modality agreement for the assessment of herniated cervical discs. MDCT images have a tendency to underestimate the anterior/posterior extent of the herniated disc compared with MRI.

Clinical utility of reflex studies in assessing cervical radiculopathy

We prospectively studied the diagnostic utility of upper limb segmental reflexes in patients with suspected cervical radiculopathy (CR). Fifty-three patients (29 men and 24 women), referred for electrodiagnostic testing, were positive for at least one of four clinical criteria for CR: abnormal (1) history, (2) motor (myotomal) examination, (3) sensory (dermatomal) examination, and (4) changes in deep tendon reflexes (DTR). All underwent electrodiagnostic assessment, needle electrode examination (NEE), specialized segmental reflexes (heteronymous and Hoffman's reflexes [H reflexes]), and neuroimaging. The clinical diagnosis was supported in all 32 patients who entered the study with two or more clinical signs for CR. Abnormal NEE was found in 90% of subjects with three clinical signs, 59% with two signs, and only 10% of those with one sign. H reflexes demonstrated a sensitivity of 72% and specificity of 85% for detection of CR and were particularly helpful when forming conclusions in the 21 subjects with only one clinical sign for CR. Specialized segmental H-reflex studies of the upper limb were as sensitive and specific as neuroimaging (magnetic resonance imaging).

Artificial lamina-assisted laminoplasty performed in seven cases

OBJECT

The authors attempted to simplify the operative approach to severe multilevel cervical spondylotic myelopathy. Seven patients with progressive and severe myelopathy underwent modified double-door laminoplasty during a 5-month period.

METHODS

The double-door laminoplasty procedure was modified by using two artificial titanium laminae obtained by simple surgical 0.5-mm Ti-mesh (rather than by bone graft or ceramic spacers). Preoperatively, gait disturbance was present in all patients with long-tract signs on neurological examination. In all cases the sagittal diameter of the cervical spinal canal was somewhat reduced (< 10 mm) by congenital stenosis, and further severe compression of the spinal cord resulted from osteophytic bars and calcified ligamenta flava at different levels. No abnormal alignment, pathological movements, or instability was present. Computerized tomography (CT) studies demonstrated severe multilevel cervical compression, and T2-weighted magnetic resonance (MR) imaging demonstrated pathological areas of hyperintensity within the spinal cord in all cases. In the initial follow-up study (range 8-12 months), the patients who underwent this procedure experienced marked improvement of gait disturbance without any significant incidence of morbidity or complications. Postoperative CT and MR imaging studies demonstrated complete spinal cord decompression and restoration of the patency of the subarachnoid spaces.

CONCLUSIONS

The proposed procedure has the advantage of achieving both an immediate stabilization of the open laminae by means of a bridgelike mechanism and protection from the possible compression of the dural sac by paravertebral muscles.

Cervical Radiculopathy: Diagnosis and Nonoperative Management

Cervical radiculopathy presents as pain in a dermatomal distribution. This frequently represents compression of an exiting cervical nerve root by either a herniated disk or a degenerative cervical spondylotic change. Most patients will improve with nonoperative treatment, and a small percentage will require further diagnostic evaluation and ultimately surgical intervention. An understanding of the normal anatomy and the pathologic changes in cervical radiculopathy will improve the understanding of diagnosis and decision making regarding nonoperative interventions. An algorithmic approach for decision making and a review of nonoperative management are presented.

Optimizing conventional cardiac MRI in the rabbit at 0.3 T

The aim of this study was to define the most efficient way of performing cardiac MRI for anatomic information in small experimental animals, using a vertical magnetic field with a strength of 0.3 T (FONAR beta-3000M). This information may be used to improve cardiac MRI in infants and small children, since the size of a rabbit is considered comparable to that of a neonate. Experimental axial cardiac MRI studies were performed in a rabbit under general anesthesia in order to study the effects on image quality of changing various imaging parameters. These are ECG-gating, number of excitations (averages), number of warp levels, echo time (TE) and repetition time (TR). The effects of changing the size of the field of view (FOV), the slice thickness and the phase-encoding direction were also studied. We found that ECG-gating was crucial and that three excitations, TE 16 ms, and 257 vertical phase-encoding warp levels were adequate. Five-millimeter slice thickness and FOV 20 cm were preferred.

Improved receiver coil for upper thoracic spine imaging in a vertical magnetic field

To improve image quality in the upper thoracic spine, an anatomically shaped copper wire loop coil, made to fit over the patient's shoulders, was constructed. The coil was permanently mounted on a foam-rubber vest to facilitate attachment to the patient. Phantom and in vivo studies of the performance of the coil in healthy volunteers showed as much as a two times greater signal-to-noise ratio relative to that of standard coils for the upper thoracic spine. In a patient with lesions in the upper thoracic cord, the coil gave better image quality in the region of interest than did the standard coils. The coil has been integrated into the authors' routine imaging equipment and has been the coil of choice for imaging of the upper thoracic spine on their 0.3-T vertical field system.