Abstract:Introduction Magnetic resonance imaging (MRI) offers non-invasive imaging method to obtain high resolution anatomic pictures of tissue and organs. However, it has extremely low sensitivity to image the bone structures which have little water content and very short T2 relaxation time . This is opposite to the problem of x-ray based computer tomography (CT), which has good capability of imaging of bone but poor capability in imaging of soft tissue. Although MRI has become a primary diagnostic imaging tool for clinical managements of various diseases and for research, it will be ideal if MRI can be used for imaging of both soft tissue and bone structure so that imaging of bone and tissue of patients can done in the same scan session and doctors can obtain important diagnostic information on bone and its tissue interface. Furthermore, the ability of imaging of bone by MRI may also open up new MRI applications, such as using MRI for positron emission tomography (PET) attenuation correction that is critical in the latest development of the combined dual-modal MRI-PET technology. Recently an ultra-short time of repetition (UTE) MRI method has been reported for its potential application of bone imaging. UTE imaging is capable of using extremely short echo time (TE) to allow for capture of signals from bone before substantial signal decay. The purposes of this study were to investigate imaging of skull using UTE MRI and to evaluate the quality of UTE skull images by comparing UTE bone images with those obtained from CT. Methods and Subjects Eight subjects (5 men, 3 female; age range from 30 to 65 years) participated in this study under the approval of Institutional Review Board. CT images were collected followed by MR scans. All CT skull images were obtained using routine brain protocol with imaging parameters of 80 kV, 190 mA, 512 x 512 matrix, and 3mm axial sections on a HiSpeed CT/iCT scanner (GE Healthcare). The MRI was performed on a 3 Tesla MR scanner (Siemens Medical Solutions) to obtain images covering the entire volume of the head. A multi-echo UTE sequence described previously was used. This sequence consists of one 60 s long non-selective RF pulse followed by a 40 s transmit/receive switch time, and a 100% asymmetric data acquisition from the center to the surface of a sphere in the k-space using a threedimensional (3D) radial sampling trajectory. The shortest TE used for this study was 0.07 ms; the second and longer TE was 2.76 ms. Other parameters included TR = 21.7 ms, flip angle = 15°, FOV = 22 × 22 cm, image matrix = 192 × 192, bandwidth = 64000 Hz/pixel, total radial projections = 40 000, radial projections per excitation = 176, numbers of average = 2. The resultant skull MR images were obtained by subtracting images from two different TEs. The skull images of both CT and MRI were reviewed and evaluated by two raters blindly. To assess the geometrical match of skull bone structures from UTE and CT images, we compared skull thicknesses measured from MR and CT images in 7 regions of interest (ROIs) from each corresponding slice of the same subject. 5 slices were selected from each subject. In addition, UTE image of cortical bone were evaluated by comparing signal intensities of the outer layer, diploe, and inter layer of the skull measured from both CT and MR images. The ratios of bone intensity of MR images to bone intensity of CT images in different ROIs were calculated. The correlation of skull thickness between MR images and CT images were analyzed using Statistical Package (SPSS 15.0) and results are used to determine the quality of UTE images of the bone. Significance was defined as P < 0.05. Results and Discussion Cortical bone images were successfully obtained from all subjects using the UTE method. Figure 1 shows two slices of brain images selected from different axial section of the head, showing cortical bone structure delineated by UTE MRI and CT. Skull structures can be clearly visualized in UTE images and matched well with that from CT. ROI analysis, as demonstrated in Figure 2, allowed us to evaluate and compare UTE bone images with CT more quantitatively. The comparison of the skull thickness measured from UTE and CT images are very close with a 10% at the directions of 3 to 10 clock in a selected slice. The skull thickness measured from UTE images showed good agreement with those obtained from CT images in different slices from the same subject as there is no statistical difference between the thickness measured from CT and UTE images (P = 0.87). There is also a good correlation between the thickness measurements obtained from CT and UTE images with R=0.97 (Figure 3), suggesting that UTE images have minimal structural distortions. Further signal intensity based evaluation showed that there is no statistical difference in the signal intensity ratio (UTE:CT) in outer, inner layer and diploe of the skull (Figure 4). This study provided a quantitative evaluation of the skull images obtained using UTE MRI and a direct comparison to the skull images from CT. Although the resolution of the skull images measured by UTE MRI is lower than those typically obtained by the CT images, skull images obtained from UTE MRI offer sufficient image quality for providing diagnostic information. The comparison of bone thickness measured from skull images by UTE MRI and those of CT suggested that UTE MRI could provide a close-match image of the cortical bone. The relatively consistent ratios of the signal intensities of UTE and CT may help the development of attenuation correction algorithms which are of interest in the applications of the combined MRI-PET system. Conclusion UTE MRI makes imaging of the short-T2 structures, such as cortical bone and skull, possible. Bone image obtained from UTE MRI enables MRI based PET attenuation correction when implemented in newly available combined MRI-PET system. With further increasing experience of improving and interpreting the bone images by UTE MRI, UTE MRI may provide a practical alternative to CT for bone imaging in some clinical situations. 2a 2b

High-Contrast Lumbar Spinal Bone Imaging Using a 3D Slab-Selective UTE Sequence

Self-Adaptive Ultrasound Scanning System for Imaging Human Spine

Editorial for "comparing CT-Like Images Based on Ultra-Short Echo Time and Gradient Echo T1-Weighted MRI Sequences for the Assessment of Vertebral Disorders Using Histology and True CT As the Reference Standard"

CT-like images based on T1 spoiled gradient-echo and ultra-short echo time MRI sequences for the assessment of vertebral fractures and degenerative bone changes of the spine

Bone Imaging of the Knee Using Short-Interval Delta Ultrashort Echo Time and Field Echo Imaging

A 3D MRI sequence for computer assisted surgery of the lumbar spine

Deep-Learning-Aided Evaluation of Spondylolysis Imaged with Ultrashort Echo Time Magnetic Resonance Imaging

Assessment and Improvement of a Novel ultrasound-based 3D Reconstruction Method: Registered for Lumbar Spine

Ultra-short TE ( UTE ) Imaging of Skull and a Quantitative Comparison of Skull Images Obtained from MRI and CT

Use of Ultra‐Short Echo Time MRI to Improve Temporal Bone Imaging

Deep learning constrained compressed sensing reconstruction improves high-resolution three-dimensional (3D) T2-weighted turbo spin echo magnetic resonance imaging (MRI) of the lumbar spine

Super-resolution deep learning reconstruction approach for enhanced visualization in lumbar spine MR bone imaging

3D balanced SSFP UTE MRI for multiple contrasts whole brain imaging

Ultrashort-Echo-Time MRI of the Disco-Vertebral Junction: Modulation of Image Contrast via Echo Subtraction and Echo Times

Three contrasts in 3 min: Rapid, high-resolution, and bone-selective UTE MRI for craniofacial imaging with automated deep-learning skull segmentation

A 3-D Ultrasound Tomography Method for Bone Morphology Evaluation

Clinical Utility of a Novel Ultrafast T2-Weighted Sequence for Spine Imaging

Rapid lumbar MRI protocol using 3D imaging and deep learning reconstruction

Fast SPECT/CT planar bone imaging enabled by deep learning enhancement

Simultaneous Quantitative Susceptibility Mapping of Articular Cartilage and Cortical Bone of Human Knee Joint Using Ultrashort Echo Time Sequences

Ultra-short Echo-time Imaging in Quantitative Assessment of Cortical Bone in Cases with Osteoporosis