MR also has been used to measure structural parameters in animal models of osteoporosis. Jiang et al. treated an ovariec- tomized sheep model of osteoporosis with salmon calcitonin, an osteoclast inhibitor, to determine if structural parameters in the neck of the femur could be maintained. It was found that BV/TV and Tb.N decreased and Tb.Sp increased in ovariectomized sheep. Structural parameters of sheep treated with salmon calcitonin were equivalent to sham operated sheep. Small-bore micro-MRI has been used to study osteo- porotic bone structure in ovariectomized rats. Analysis of MR images revealed differences in osteoporotic trabecular structure that DXA could not detect.

Takahashi et al. have investigated the effects of corticosteroid on bone structure in rabbit femurs using magnetic resonance microimaging (|iMRI). They found that short term, high doses of corticosteroids resulted in a decrease in trabecular bone volume through trabecular thinning with little change in trabecular network, trabecular number or trabecular spacing. Using MR spectroscopy they also determined that hematopoietic bone marrow was converted to fatty marrow in rabbits treated with corticosteroid.

Read the rest of this entry »

Several studies have explored how MR images compare with other imaging modalities in determining structural parameters (Table I). Hipp et al. and Hopper et al. used small-bore MRI with resolutions of 92x92x92 |im³ and 23x23x39 |im³ respec­tively. All other studies were performed on 1.5 or 3T scanners with in-plane resolution of 100-150 mm and a slice thickness of 300 |im on in vitro bone cubes. Weber et al. compared MR in vivo and in vitro trabecular bone images from mice with histological sections. They found parame­ters derived from in vivo images correlated better with histo­logical parameters than did in vitro images and attributed the difference to the better MR signal from bone marrow than for­malin. These studies indicate that MR derived architectural parameters correlate well with measures taken at much high­er resolutions. In general, MR tended to overestimate BV/TV and Tb.Th and underestimate Tb.Sp due to partial volume ef­fects.

Architechtural parameters have also been compared to bone mineral density (BMD) and mechanical strength in the radius, lumbar vertebrae, femur, calcaneus and among various sights. In these studies correlations coeffi­cients for BV/TV, Tb.Th, and Tb.N with BMD or mechanical strength were between 0.5 and 0.8. All studies found that Tb.Sp had a correlation coefficient with BMD or mechanical strength of -0.5 to -0.6, indicating that the spacing between the trabeculae increases as BMD and mechanical strength de­crease. Studies also found that combining BMD and trabecular structural parameters improved correlations with mechani­cal strength.

Read the rest of this entry »

Bone mineral density and trabecular structure together deter­mine the mechanical strength of trabecular bone. The main ob­jective of imaging trabecular bone structure is to determine morphological parameters of the trabecular architecture. These morphologic parameters may help to determine the efficacy of therapeutic treatments for osteoporosis and predict individuals at risk for bone fracture. Standard histomorphometric mea­sures of bone structure include: bone volume fraction (BV/TV), trabecular thickness (Tb.Th), mean intercept lenght, trabecular number (Tb.N), and trabecular spacing (Tb.S). These parame­ters have been adapted to analyze MR images of trabecular structure.

Because the resolution of in vivo MR images is on the same scale as trabecular dimensions, these histomorphometric para­meters are the measures of the trabeculae projected across the slice thickness. Majumdar et al. introduced “apparent” mea­sures, indicating that the morphometric measures obtained from in vivo MR images may not be exactly equivalent, howev­er are related to those obtained from higher resolution modali­ties. It was found that trabecular spacing and trabecular number are relatively independent of resolution. Trabecu- lar thickness, however, was strongly dependent on resolution with lower resolutions resulting in thicker trabeculae. A 3 dimensional distance technique was introduced by Hilde- brand and Ruegsegger to determine mean thickness by fitting spheres within the structure. This measure was able to distinguish between trabecular bone composed of a greater percentage of plates or rods. It has also been used calcu­late histomorphometric parameters such as app.Tb.Th and app.Tb.Sp from MR images. The morphological para­meters calculated using the distance technique correlated well with those calculated using the mean intercept length. Because osteoporosis is thought to result in a thinning of tra- beculae and loss of trabecular connectivity, measures of con­nectivity are important in determining osteoporotic bone quali­ty. Connectivity measures have been established to measure the degree of connectivity of the trabecular network in trabecu­lar bone. Connectivity indicates the maximum number of branches that can be broken before the structure is separat­ed into two parts. It is a topological invariant, which means it does not change if the structure is stretched, bent, twisted or other rubber-like deformation. Connectivity can be calculated in terms of the Euler characteristic. Previous studies have used the Euler number to analyze MR images of trabecular bone and found that connectivity can vary between regions within a bone and is significantly correlated with bone density and bone volume fraction.

Read the rest of this entry »

After obtaining an MR image, pre-processing of the image is usually required in order to improve the signal-to-noise ratio and image quality and make it possible to differentiate marrow from bone trabeculae. Pre-processing may include coil correc­tion, noise reduction, motion correction, and thresholding. Coil correction is required to correct spatial variations in the sensi­tivity of the detection coil as tissue close to the coil usually ap­pears brighter than tissue further away from the coil. Coil cor­rection algorithms depend on the structure of the specific coil. Coils that completely surround the object being scanned (e.g. bird-cage coil) provide sufficient in-plane homogeneity, making longitudinal correction sufficient. In surface coils, which may not provide in-plane homogeneity, a low-pass filter (LPF)- based coil correction scheme is necessary (Figure 4). Noise reduction improves the signal-to-noise ratio and may be accomplished using a median low pass filter, in which the median of the pixels in a certain kernel size (e.g. 3×3 pixels) surrounding a pixel becomes the new filtered value for the pixel. A low pass filter removes high signal noise, while preserving the low signal data. The kernel median allows edge detection, whereas the kernel mean would smooth the data and blur the edges. Hwang et al. proposed a histogram deconvolution method in order to obtain a noiseless his­togram for trabecular bone. In this method a probability distribution of the noise (e.g. Gaussian) and an initial esti­mate of the noiseless histogram are assumed in order to pre­dict a histogram. The predicted histogram is iteratively improved by comparing it to the measured histogram. The noiseless histogram and raw image are used to produce a noiseless image. Others have proposed wavelet-based thresholding that allows more local noise reduction while re­taining relevant detail information. Imaging trabeculae on the order of 100 ^m means that a small amount of motion will affect the image. Various techniques have been devised to correct for motion artifacts. Navigator correction alters the echo sequence, adding echos to sense small displacements. The data is corrected in k-space by analyzing the phase shift and adjusting for translational mo­tions. Studies have shown that navigator correction improves reproducibility and accuracy of trabecular bone parameters. Retrospective motion correction can also be performed with autofocusing (Figure 5). This technique applies trial phase shifts to the data and compares the resulting image with the original. An entropy focusing criterion is applied to minimize the amount of entropy in the image and obtain maxi­mum contrast.

Perhaps the most critical pre-processing step is thresholding, which allows delineation of the trabeculae and the marrow. Be­cause the resolution of in vivo MR images is on the same scale as the trabecular width, partial volume effects occur. In partial voluming, a single voxel may contain signals from multiple tis­sue types. The voxel intensity is the average signal from the various tissues. The histogram of trabecular bone, therefore, is not bi-modal with marrow and bone peaks, but rather mono modal with a peak intensity between the values of marrow and bone. Various thresholding methods have been established in order to segment the bone from the marrow where partial vol­ume effects are an issue. Majumdar et al. proposed a dual thresholding method in which the threshold for bone was a mean pixel value taken in the cortical shell and the threshold for marrow was the lower signal intensity at which the his­togram reached half its peak.

Read the rest of this entry »

MRI basics

Nuclei with an odd number of protons and neutrons (such as hydrogen) have a magnetic moment causing the nucleus to act like a small magnet in the presence of an external magnetic field. The magnetic field of the nucleus aligns in the direction of the external magnetic field. Magnetic resonance imaging uses radio frequency (RF) pulses in a magnetic field in order to alter the spin of protons in the tissue. Coils detect the change in net magnetization, which after mathematical reconstruction provides spatial and compositional information of the tissue being imaged. Because clinical MRI usually detects magnetization of hydrogen, compositional information is limited to molecules containing hydrogen, such as water, body fat, and cholesterol. In a MRI scanner, proton spins in the body align in the direction of the external magnetic field. When an RF pulse is applied, the proton spins change, altering the magnetization. The time it takes for the spin to regain its alignment with the external mag­netic field after the RF pulse is turned off depends on the mole­cule (size and structure) and its surroundings. By altering the se­quence of the RF pulses and the gradient of the magnetic field, the location and type of tissue being imaged can be controlled. The signal received in an MR image reflects intrinsic factors of the tissue, either spin density or relaxation properties of the nu­clei. Spin-lattice relaxation time (T1) is the time it takes a tissue to regain longitudinal magnetization after a 90° RF pulse makes the spins perpendicular to the external magnetic field. T1 is a measure of energy transfer to the surroundings (lattice) as the proton recovers its normal spin. T1 relaxation times gen­erally are between 300-2000 msec. Spin-spin relaxation time (T2) is a measure of how long the proton spins remain in phase after an RF pulse. Interaction with other molecules (e.g. diffusion) affects the T2 relaxation time. As natural motion of the proton increases, such as in liquids, T2 increases. Water, therefore, has a long T2, and appears white in T2-weighted im­ages. T2 relaxation times are shorter than T1 and can range from 30-150 msec. Inhomogeneities in the magnetic field can also affect T2. A static internal field (caused by large, slow- moving proteins or rigid trabeculae for example), may addition­ally alter the local magnetic environment and affect T2. T2* combines the effects of molecular interactions (T2) and these field inhomogeneities. In addition to relaxation times, more complicated measures may also be obtained from the MRI sig­nal, such as phase analysis, relaxation time distribution, and chemical composition. MR images also can reflect the behavior of water or fat alone. Figure 2 shows a radiograph of a proxi­mal femur and a comparative fat suppressed MR image.

Read the rest of this entry »

Introduction

Osteoporosis is a metabolic disorder that results in a decrease in bone mineral density and an alteration in the trabecular ar­chitectural structure. Osteoporotic bone has decreased me­chanical strength making it prone to fracture, especially atrau­matic vertebral fractures and fall-related hip and radius frac­tures. Osteoporosis is clinically diagnosed using measurement of bone mineral density. Bone mineral density is usually mea­sured using x-ray or ultrasound imaging techniques. In x-ray imaging (such as dual energy x-ray absoptiometry, DEXA, and quantitative computer tomography, QCT) the image intensity relates to the tissue mineral density. In ultrasound, image in­tensity reflects the change in frequency and amplitude of the sound wave traveling through the tissue. X-ray techniques use ionizing radiation, which can have deleterious effects in suffi­cient doses. Ultrasound, though harmless, provides only a small field of view, which may limit the accuracy of the mea­surement. In addition to bone density, the quality of bone which includes bone micro-architecture is of interest. Recent ad­vances in micro-computed tomography, a x-ray based 3D tech­nique has made it possible to obtain images of trabecular bone micro-architecture. Read the rest of this entry »