Armando Manduca

Highly Undersampled Magnetic Resonance Image Reconstruction via Homotopic

In clinical magnetic resonance imaging (MRI), any reduction in scan time offers a number of potential benefits ranging from high-temporal-rate observation of physiological processes to improvements in patient comfort. Following recent developments in compressive sensing (CS) theory, several authors have demonstrated that certain classes of MR images which possess sparse representations in some transform domain can be accurately reconstructed from very highly undersampled K-space data by solving a convex lscr1-minimization problem. Although lscr1-based techniques are extremely powerful, they inherently require a degree of over-sampling above the theoretical minimum sampling rate to guarantee that exact reconstruction can be achieved. In this paper, we propose a generalization of the CS paradigm based on homotopic approximation of the lscr0 quasi-norm and show how MR image reconstruction can be pushed even further below the Nyquist limit and significantly closer to the theoretical bound. Following a brief review of standard CS methods and the developed theoretical extensions, several example MRI reconstructions from highly undersampled K-space data are presented.

\ell_{0}

The well-documented effectiveness of palpation as a diagnostic technique for detecting cancer and other diseases has provided motivation for developing imaging techniques for noninvasively evaluating the mechanical properties of tissue. A recently described approach for elasticity imaging, using propagating acoustic shear waves and phase-contrast MRI, has been called magnetic resonance elastography (MRE). The purpose of this work was to conduct preliminary studies to define methods for using MRE as a tool for addressing the paucity of quantitative tissue mechanical property data in the literature. Fresh animal liver and kidney tissue specimens were evaluated with MRE at multiple shear wave frequencies. The influence of specimen temperature and orientation on measurements of stiffness was studied in skeletal muscle. The results demonstrated that all of the materials tested (liver, kidney, muscle and tissue-simulating gel) exhibit systematic dependence of shear stiffness on shear rate. These data are consistent with a viscoelastic model of tissue mechanical properties, allowing calculation of two independent tissue properties from multiple-frequency MRE data: shear modulus and shear viscosity. The shear stiffness of tissue can be substantially affected by specimen temperature. The results also demonstrated evidence of shear anisotropy in skeletal muscle but not liver tissue. The measured shear stiffness in skeletal muscle was found to depend on both the direction of propagation and polarization of the shear waves.

-Minimization

The purpose of this study was to obtain normative data using magnetic resonance elastography (MRE) (a) to obtain estimates of the shear modulus of human cerebral tissue in vivo and (b) to assess a possible age dependence of the shear modulus of cerebral tissue in healthy adult volunteers. MR elastography studies were performed on tissue-simulating gelatin phantoms and 25 healthy adult volunteers. The data were analyzed using spatiotemporal filters and a local frequency estimating algorithm. Statistical analysis was performed using a paired t-test. The mean shear stiffness of cerebral white matter was 13.6 kPa (95% CI 12.3 to 14.8 kPa); while that of gray matter was lower at 5.22 kPa (95% CI 4.76 to 5.66 kPa). The difference was statistically significant (p<0.0001).

Tissue characterization using magnetic resonance elastography: preliminary results*

While the contractile properties of skeletal muscle have been studied extensively, relatively little is known about the elastic properties of muscle in vivo. Magnetic resonance elastography (MRE) is a phase contrast‐based method for observing shear waves propagating in a material to determine its stiffness. In this work, MRE is applied to skeletal muscle under load to quantify the change in stiffness with loading. A mathematical model of muscle is developed that predicts a linear relationship between shear stiffness and muscle load. The MRE technique was applied to bovine muscle specimens (N = 10) and human biceps brachii in vivo (N = 5). Muscle stiffness increased linearly for both passive tension (14.5 ± 1.77 kPa/kg) and active tension, in which the increase in stiffness was dependent upon muscle size, as predicted by the model. A means of noninvasively assessing the viscoelastic pro‐perties of skeletal muscle in vivo may provide a useful method for studying muscle biomechanics in health and disease. J. Magn. Reson. Imaging 2001;13:269–276.

Magnetic Resonance Elastography of the Brain

Dynamic magnetic resonance elastography can visualize and measure propagating shear waves in tissue-like materials subjected to harmonic mechanical excitation. This allows the calculation of local values of material parameters such as shear modulus and attenuation. Various inversion algorithms to perform such calculations have been proposed, but they are sensitive to areas of low displacement amplitude (and hence low SNR) that result from interference patterns due to reflection and refraction. A spatio-temporal directional filter applied as a pre-processing step can separate interfering waves so they can be processed separately. Weighted combinations of inversions from such directionally separated data sets can significantly improve reconstructions of shear modulus and attenuation.

Magnetic resonance elastography of skeletal muscle.

OBJECTIVE Liver stiffness is associated with portal hypertension in patients with chronic liver disease. However, the relation between spleen stiffness and clinically significant portal hypertension remains unknown. The purposes of this study were to determine the feasibility of measuring spleen stiffness with MR elastography and to prospectively test the technique in healthy volunteers and in patients with compensated liver disease. MATERIALS AND METHODS Spleen stiffness was measured with MR elastography in 12 healthy volunteers (mean age, 37 years; range, 25-82 years) and 38 patients (mean age, 56 years; range, 36-60 years) with chronic liver disease of various causes. For patients with liver disease, laboratory findings, spleen size, presence and size of esophageal varices, and liver histologic results were recorded. Statistical analyses were performed to assess all measurements. RESULTS MR elastography of the spleen was successfully performed on all volunteers and patients. The mean spleen stiffness was significantly lower in the volunteers (mean, 3.6 +/- 0.3 kPa) than in the patients with liver fibrosis (mean, 5.6 +/- 5.0 kPa; range, 2.7-19.2 kPa; p < 0.001). In addition, a significant correlation was observed between liver stiffness and spleen stiffness for the entire cohort (r(2) = 0.75; p < 0.001). Predictors of spleen stiffness were splenomegaly, spleen volume, and platelet count. A mean spleen stiffness of 10.5 kPa or greater was identified in all patients with esophageal varices. CONCLUSION MR elastography of the spleen is feasible and shows promise as a quantitative method for predicting the presence of esophageal varices in patients with advanced hepatic fibrosis.

Spatio-temporal directional filtering for improved inversion of MR elastography images

The purposes of this study were to develop a method to measure brain and white matter hyperintensity (leukoaraiosis) volume that is based on the segmentation of the intensity histogram of fluid‐attenuated inversion recovery (FLAIR) images and to assess the accuracy and reproducibility of the method. Whole‐head synthetic image phantoms with manually introduced leukoaraiosis lesions of varying severity were constructed. These synthetic image phantom sets incorporated image contrast and anatomic features that mimicked leukoaraiosis found in real life. One set of synthetic image phantoms was used to develop the segmentation algorithm (FLAIR‐histoseg). A second set was used to measure its accuracy. Test retest reproducibility was assessed in 10 elderly volunteers who were imaged twice. The mean absolute error of the FLAIR‐histoseg method was 6.6% for measurement of leukoaraiosis volume and 1.4% for brain volume. The mean test retest coefficient of variation was 1.4% for leukoaraiosis volume and 0.3% for brain volume. We conclude that the FLAIR‐histoseg method is an accurate and reproducible method for measuring leukoaraiosis and whole‐brain volume in elderly subjects. J. Magn. Reson. Imaging 2001;14:668–676.

Feasibility of In Vivo MR Elastographic Splenic Stiffness Measurements in the Assessment of Portal Hypertension

To test patient acceptance and reproducibility of the 3D magnetic resonance elastography (MRE) brain exam using a soft vibration source, and to determine if MRE could noninvasively measure a change in the elastic properties of the brain parenchyma due to Alzheimers disease (AD).

FLAIR Histogram Segmentation for Measurement of Leukoaraiosis Volume

Magnetic resonance elastography (MRE) is a technique for quantifying material properties by measuring cyclic displacements of propagating shear waves. As an alternative to dynamic harmonic wave MRE or quasi‐steady‐state methods, the idea of using a transient impulse for mechanical excitation is introduced. Two processing methods to calculate shear stiffness from transient data were developed. The techniques were tested in phantom studies, and the transient results were found to be comparable to the harmonic wave results. Transient wave based analysis was applied to the brains of six healthy volunteers in order to assess the method in areas of complex wave patterns and geometry. The results demonstrated the feasibility of measuring brain stiffness in vivo using a transient mechanical excitation. Transient and harmonic methods both measure white matter (∼12 kPa) to be stiffer than gray matter (∼8 kPa). There were some anatomic differences between harmonic and transient MRE, specifically where the transient results better depicted the deeper structures of the brain. Magn Reson Med 53:628–639, 2005.

Decreased brain stiffness in Alzheimer's disease determined by magnetic resonance elastography

PURPOSE To investigate a novel locally adaptive projection space denoising algorithm for low-dose CT data. METHODS The denoising algorithm is based on bilateral filtering, which smooths values using a weighted average in a local neighborhood, with weights determined according to both spatial proximity and intensity similarity between the center pixel and the neighboring pixels. This filtering is locally adaptive and can preserve important edge information in the sinogram, thus maintaining high spatial resolution. A CT noise model that takes into account the bowtie filter and patient-specific automatic exposure control effects is also incorporated into the denoising process. The authors evaluated the noise-resolution properties of bilateral filtering incorporating such a CT noise model in phantom studies and preliminary patient studies with contrast-enhanced abdominal CT exams. RESULTS On a thin wire phantom, the noise-resolution properties were significantly improved with the denoising algorithm compared to commercial reconstruction kernels. The noise-resolution properties on low-dose (40 mA s) data after denoising approximated those of conventional reconstructions at twice the dose level. A separate contrast plate phantom showed improved depiction of low-contrast plates with the denoising algorithm over conventional reconstructions when noise levels were matched. Similar improvement in noise-resolution properties was found on CT colonography data and on five abdominal low-energy (80 kV) CT exams. In each abdominal case, a board-certified subspecialized radiologist rated the denoised 80 kV images markedly superior in image quality compared to the commercially available reconstructions, and denoising improved the image quality to the point where the 80 kV images alone were considered to be of diagnostic quality. CONCLUSIONS The results demonstrate that bilateral filtering incorporating a CT noise model can achieve a significantly better noise-resolution trade-off than a series of commercial reconstruction kernels. This improvement in noise-resolution properties can be used for improving image quality in CT and can be translated into substantial dose reduction.