Jianqi Li¹, Tian Liu^2,3, Shixin Chang⁴, Fang Dong¹, Hongwei Jiang¹, and Yi Wang^2,3
1Shanghai Key Laboratory of Magnetic Resonance and Department of Physics, East China Normal University, Shanghai, Shanghai, China, ²Department of Biomedical Engineering, Cornell University, Ithaca, New York, United States, ³Department of Radiology, Weill Medical College of Cornell University, New York, United States, ⁴Department of Radiology, Yueyang Hospital of Integrated Traditional Chinese and Western Medicine, Shanghai, China

Multi-echo GRE sequences using bipolar readout gradients can offer efficient data acquisition for quantitative susceptibility mapping (QSM), but the bipolar gradient approach requires correction for phase errors. A simple linear phase correction in read-out direction is demonstrated in this study to allow accurate QSM in human brain using the bipolar multiecho GRE sequence. For all subjects and all ROIs studies, the bipolar multiecho acquisition provides good quantitative agreement with and reduces the noise in the unipolar acquisition.

Alexey V. Dimov^1,2, Tian Liu³, Pascal Spincemaille², Jacob S. Ecanow^4,5, Huan Tan⁶, Robert R. Edelman^4,7, and Yi Wang^1,2
1Biomedical Engineering, Cornell University, Ithaca, NY, United States, ²Radiology, Weill Cornell Medical College, New York, NY, United States,³MedImageMetric LLC, New York, NY, United States, ⁴Radiology, NorthShore University HealthSystem, Chicago, IL, United States, ⁵Radiology, University of Chicago Pritzker School of Medicine, Chicago, IL, United States, ⁶Surgery (Neurosurgery), University of Chicago, Chicago, IL, United States,⁷Radiology, Northwestern University Feinberg School of Medicine, Chicago, IL, United States

Iterative decomposition of water and fat with echo asymmetry and least squares estimation (IDEAL) method can be used for estimation of the field distribution within the tissue of interest, which then may be processed for susceptibility estimation. However, errors in the assumed model of chemical spectrum of fat will propagate to the final susceptibility map as streaking artifacts. In the current research we propose an iterative algorithm that performs adjustment of the chemical shift value based on detected inconsistencies between estimated tissue structure and field map.

Samir D. Sharma¹, Diego Hernando¹, and Scott B. Reeder^1,2
1Radiology, University of Wisconsin - Madison, Madison, WI, United States, ²Medical Physics, University of Wisconsin - Madison, Madison, Wisconsin, United States

Precise estimation of the B0 field map is a critical component of quantitative susceptibility mapping (QSM). Using QSM, it may be possible to estimate the liver iron concentration (LIC) from the B0 field map. However, in order for this B0-based measurement of LIC to be precise (i.e. have low standard deviation), the estimate of the B0 field itself must be precise. In this work, we have analyzed the theoretical effects of the echo time sampling on the precision of the B0 field map estimate, and further, translated these results into precision limits of the LIC estimate.

Sherman Xuegang Xin^1,2 and Fanrui Fu^1
1Biomedical Engineering, Southern Medical University, Guangzhou, Guangdong, China, ²Bernard and Irene Schwartz Center for Biomedical Imaging (CBI), New York University School of Medicine, New York, New York, United States

Due to the strong dependence of tissue electrical properties on temperature, it is important to consider the potential effects of intense heating on the electromagnetic fields during MRI, as in MR-guided Focused UltraSound. Changes in tissue electrical properties during an exam may affect both efficacy of pulses designed with knowledge of B1 field distributions and the SAR pattern. Here the temperature- and frequency-dependent electric properties of porcine tissues were measured ex vivo. The obtained data can be used to calculate the electromagnetic field and SAR distribution in and around a region of heating in the MR environment.

Necip Gurler¹, Omer Faruk Oran¹, and Yusuf Ziya Ider^1
1Electrical and Electronics Engineering, Bilkent University, Ankara, Ankara, Turkey

In this study, a new convection reaction equation based formula to reconstruct the electrical properties (EPs) using B1- data is derived. Both simulation and experimental data are presented. Proposed method is applicable to a standard MR quadrature body coil and a receive phased array coil.

Woo Chul Jeong¹, Zi Jun Meng¹, Munish Chauhan¹, Hyung Joong Kim¹, Oh In Kwon², and Eung Je Woo^1
1Kyung Hee University, Yongin, Korea, ²Konkuk University, Seoul, Korea

The electrical conductivity of human tissues has been inferred from in vitro and/or ex vivo measurements of different species. However, in vivo tissues may differ from in vitro or ex vivo state due to the complicated tissue responses in living organs. In this study, we performed in vivo MREIT imaging of human lower extremity and compared the resulting conductivity images with ex vivo biological tissue phantom images. Comparing with phantom results, the human conductivity images showed similar pattern except muscle and adipose tissue due to the anisotropic characteristic of muscle and irrigation of high conductive fluids in the adipose tissue.

Jaewook Shin¹, Joonsung Lee^2,3, Min-oh Kim¹, Narae Choi¹, and Dong-Hyun Kim^1
1Department of Electrical and Electronic Engineering, Yonsei University, Seodaemun-Gu, Seoul, Korea, ²SIRIC, Yonsei University, Seodaemun-Gu, Seoul, Korea, ³Nanomedical Research Center, Yonsei University, Seoul, Korea

Phase-based EPT is efficient method to measure in-vivo electrical conductivity distribution using only B1 phase information. To measure accurate conductivity value using phase-based EPT, spatial variation of the magnitude of transmit (B1+) and receive (B1-) field has to be negligible.However, when signal was received using multi-Rx, the spatial variation of B1- magnitude is non-negligible. In this study, we suggest a patient-individual coil combine method using two-stage optimization.

Omer Faruk Oran¹, Necip Gurler¹, and Yusuf Ziya Ider^1
1Electrical and Electronics Engineering, Bilkent University, Ankara, Turkey

For MREPT, apart from various assumptions made for deriving algorithms, one of the major problem is still reconstructing the electrical properties (EP) at the tissue boundaries where gradients of these variables are high (non-constant region). In this study, we propose an easily implementable point-wise formula for reconstructing the EPs in all regions including transition regions. The algorithm is tested with noisy and noiseless simulation data and successful reconstructions are obtained. The proposed algorithm is also suitable for practically reconstructing EPs when transmit and/or receive arrays are used.

Jakob Meineke¹ and Ulrich Katscher^1
1Philips Research Europe, Hamburg, Germany

Phase-based Electric Properties Tomography (EPT) can be used to measure the conductivity of tissue in vivo. However, the method is limited by the assumption that both receive (RX) and transmit (TX) radiofrequency fields are spatially homogenous. Here, it is shown that the systematic errors of phase-based EPT can be reduced by using a local, offline shimming procedure. It utilizes B1-mapping and the additional degrees of freedom present in multi-TX/RX systems. A primary goal of the technique is to improve conductivity imaging of breast tumors.

Edmond Balidemaj¹, Johan Trinks¹, Rob F. Remis², Hans Crezee¹, Astrid van Lier³, Aart J. Nederveen⁴, and Cornelis A.T. van den Berg^3
1Radiotherapy, Academic Medical Center, Amsterdam, Netherlands, ²Technical University Delft, Delft, Netherlands, ³UMC Utrecht, Utrecht, Netherlands, ⁴Academic Medical Center, Amsterdam, Netherlands

In this work we present a regularized CSI-EPT method. Furthermore, we extend the reconstruction algorithm such that multiple B1+ data sets for different shim (antenna phase) settings can be simultaneously included in the iterative CSI-EPT process leading to an overall improvement of the reconstructed dielectric values.Electric field and dielectric tissue property maps where retrieved for in vivo simulations and phantom experiments.

Jae Mo Park¹, Mohammad Mehdi Khalighi², Deqiang Qiu³, Greg Zaharchuk¹, Chunlei Liu⁴, and Michael Moseley^1
1Radiology, Stanford University, Stanford, CA, United States, ²Applied Science Laboratory, GE Healthcare, CA, United States, ³Radiology, Emory University, GA, United States, ⁴Radiology, Duke University, Durham, NC, United States

Recently, the multi-pole based (p-space) susceptibility tensor imaging (STI) that only requires single gradient echo MR scan was developed and was successful to produce a tensor image of mouse brain at 7T, which was strikingly correspondence to diffusion tensor image. In this study, we compared the multi-pole based STI at 3T and 7T in a human brain, and explored the T2* dependence of the susceptibility anisotropy as well as the k-space truncation effects caused by shifts of k-space .

Saurav ZK Sajib¹, Woo Chul Jeong¹, Hyung Joong Kim¹, Oh In Kwon², and Eung Je Woo^1
1Kyung Hee University, Yongin, Korea, ²Konkuk University, Seoul, Korea

Multi-frequency spectra of electrical conductivity can provide new information for the physiological and pathological status of biological tissues. Recent simultaneous dual-frequency conductivity imaging by the combination of MREIT and MREPT can provide the different kind of internal conductivity distributions at low frequency (below 1 kHz) induced by externally injected currents and at the Larmor frequency related to the strength of the magnetic field, respectively. Despite of the potential to describe the membrane properties and spectral information, MREPT and MREIT techniques still suffer from the inherent weak signals and noise amplification by the differentiation of the measured phase data.

Sung-Hong Park¹ and Seong-Gi Kim^2,3
1Bio and Brain Engineering, Korea Advanced Institute of Science and Technology, Daejeon, Daejeon, Korea, ²Radiology, University of Pittsburgh, Pittsburgh, PA, United States, ³Center for Neuroscience Imaging Research, Institute for Basic Science (IBS), Dept. of Biological Sciences, SKKU, Suwon, Korea

We proposed a new approach to simultaneous measurement of oxygen saturation levels and diameter of a single intracortical vein of rat brain at 9.4T. The method is based on minimization of error between complex MR signals from experiments and simulations. It does not limit physical angle or location of veins. The MION-based phantom study showed measured susceptibility and diameter within 10% of error, whereas in vivo oxygen saturation levels and diameters were over-estimated. Visually the complex signal distributions from the experiment and simulations were very close to each other. The method is promising but requires further studies to improve accuracy.

Jean-Christophe Brisset¹, Pippa Storey¹, and Yulin Ge^1
1Radiology, New York University Langone, New York, NY, United States

Susceptibility blooming effect in MRI refers to signal loss due to local field inhomogeneity and spin dephasing that extends outside the true object size with enhanced visibility. The purpose of this study is to explore and quantify the spatial resolution dependence of blooming factor at 7T. A phantom was made using plastic cylinders (diameter 3mm) filled with different iron concentrations of Ferumoxytol. Our data suggest that the acquisition matrix have an impact to measure the susceptibility blooming effect. In a clinical perspective, the blooming factor could be helpful to enhance the detection of smaller lesions (e.g. those smaller than the voxel).

Sarah Eskreis-Winkler¹, Kofi Deh¹, Cynthia Wisnieff², Pascal Spincemaille¹, Tian Liu³, Ryan Brown⁴, Moonsoo Jin⁵, and Yi Wang^1,2
1Weill Cornell Medical College, New York, New York, United States, ²Bioengineering, Cornell University, New York, New York, United States,³MedImageMetric, LLC, NY, United States, ⁴Radiology, NYU Langone Medical Center, NY, United States, ⁵Bioengineering, Cornell University, NY, United States

In this work we present phantom validation of QSM’s ability to distinguish between solid and hollow spheres of susceptibility. Since these geometries give rise to similar field patterns, they can be technically challenging to reconstruct. They are present in MS lesions and hemorrhage, among other conditions, and are thus clinically important to investigate.

Adam M Winchell^1,2, Ruitian Song¹, Ralf B Loeffler¹, Winfred Wang³, Jane Hankins³, Kathleen J Helton¹, and Claudia M Hillenbrand^1
1Radiological Sciences, St. Jude Children’s Research Hospital, Memphis, TN, United States, ²Biomedical Engineering, University of Memphis, Memphis, TN, United States, ³Hematology, St. Jude Children’s Research Hospital, Memphis, TN, United States

The most devastating complication in children with sickle cell disease (SCD) is stroke. Susceptibility-weighted imaging (SWI) with MR angiography can be used to investigate the integrity of the venous and arterial cerebrovascular system. However, some irregular or tortuous arteries also appearing hypointense due to failure of flow compensation in GRE sequence have been reported in SCD patients. The mixture of decreased venous conspicuity and arterial vessels could lead to misinterpretation or concealment of venous vasculature abnormalities. In this study, we propose a post-processing technique to remove arterial vessel contamination in SWI exams.

Ryan Hutten¹, Nisa Desai¹, Demetrius Maraganore^2,3, Robert R. Edelman^1,4, and Ying Wu^1,5
1Radiology, Northshore University Health System, Evanston, IL, United States, ²Northshore University Health System, IL, United States, ³Neurology, University of Chicago, IL, United States, ⁴Northwestern University Feinberg School of Medicine, IL, United States, ⁵Radiology, University of Chicago, IL, United States

Sensitive and reliable measurements of the substantia nigra (SN) are imperative for early detection and follow-up of Parkinson’s Disease (PD) progression. Diffusion Tensor Imaging (DTI), Magnetic Transfer Ratio (MTR) and Quantitative Susceptibility Mapping (QSM) are advanced MR modalities that have shown considerable clinical utility in PD. However these methods require labor intensive and error prone manual outlining of SN to derive quantitative measurements. Commonly used automated segmentation algorithms are currently unable to isolate the SN. We report an automated segmentation of SN and the significantly improved reliability of multiparametric MR measurements.

Haitao Zhu¹, Binbin Nie¹, Hua Liu¹, Baoci Shan¹, and Hua Guo^2
1Centre for Technology R&D, Institute of High Energy Physics, Chinese Academy of Sciences, Beijing, Beijing, China, ²Center for Biomedical Imaging Research, Department of Biomedical Engineering, School of Medicine, Tsinghua University, Beijing, China

Gisela E Hagberg^1,2 and Klaus Scheffler^1,2
1Biomedical Magnetic Resonance, University Hospital Tuebingen, Tuebingen, Germany, Germany, ²High Field Magnetic Resonance, MPI for Biological Cybernetics, Tuebingen, BW, Germany

The performance of two model-free, high-pass filtering methods for data cleaning of multi-echo GRE data at 9.4T was assessed. Along with the widely used Gaussian kernel, we explored the use of a smoothing spline function that adapts its shape to the actual k-space appearance of the MR signal. The spline-based, but not the Gaussian kernel, mirrored the temporal evolution of the MR signal with echo time and hereby mitigates strong Bo inhomogeneities, while desired phase features were maintained.

Simon Daniel Robinson¹, Barbara Dymerska¹, and Siegfried Trattnig^1
1Department of Biomedical Imaging und Image-guided Therapy, Medical University of Vienna, Vienna, Vienna, Austria

Wraps in phase images can be removed by spatial unwrapping in 2D or 3D. Generally, 2D is more prone to errors and 3D is very time consuming. 2D phase unwrapping is nearly universally performed in the axial plane. We investigate the possibility of increasing the accuracy of 2D phase unwrapping by unwrapping in other planes, and by combining the results of unwrapping in the three primary planes - Triplanar unwrapping. Results from simulated and high resolution in-vivo 7T data show that unwrapping in the sagittal plane is superior to axial, and the Triplanar approach reduces errrors to close to zero.

Joseph Dagher¹, Kambiz Nael², Arthur Gmitro³, and Ali Bilgin^4
1Medical Imaging, Electrical and Computer Engineering, University of Arizona, Tucson, AZ, United States, ²Medical Imaging, University of Arizona, Arizona, United States, ³Medical Imaging, Optical Sciences, University of Arizona, Arizona, United States, ⁴Biomedical Engineering, Electrical and Computer Engineering, University of Arizona, Arizona, United States

Combining phase images from multiple channels is a difficult problem hindered by challenges such as phase wrapping, noise and the unknown phase offset between each of the channels. We present here a method which reconstructs the underlying object’s phase image (i.e., tissue phase), as well as the channels’ phase offsets, from a single Multi-Echo Gradient Echo (MEGE) scan, without requiring a reference scan.

Sarah Eskreis-Winkler¹, Pascal Spincemaille², Dong Zhou², Tian Liu³, and Yi Wang^2
1Radiology, Weill Cornell Medical College, New York, NEW YORK, United States, ²Weill Cornell Medical College, New York, New York, United States,³MedImageMetric, LLC, New York, New York, United States

Background field removal is a key step in the QSM algorithm. Current methods perform well in the interior of the ROI, but can be corrupted by error at the boundary. To improve local field estimation at the boundary, we present iPDF, an iterative version of Project onto Dipole Fields (PDF). In our phantom data, iPDF shows reduced error at the boundary. Similar patterns are seen in vivo.

Wei Li^1,2, Bing Wu^1,3, and Chunlei Liu^1,4
1Brain Imaging and Analysis Center, Duke University, Durham, NC, United States, ²Research Imaging Institute, University of Texas Health Science Center at San Antonio, San Antonio, TX, United States, ³GE Healthcare, Beijing, China, ⁴Radiology, Duke University, Durham, NC, United States

Quantitative susceptibility mapping (QSM) and susceptibility tensor imaging (STI) are two recently developed imaging methods, which provide quantitative information of tissue chemical composition, especially myelin and iron, exquisite anatomic details, and unique information of white matter microstructures of the brain. Both QSM and STI require sophisticated steps for 3D phase unwrapping, background phase removal, and dipole inversion to derive magnetic susceptibility or susceptibility tensors. Here, we provided our methods for phase preprocessing, QSM, and STI, and its related graphical user interfaces as a Matlab-based software package, named as “STI Suite”, for free academic use.

Rei MOMOSAWA¹, Junki SAKURAI¹, and Etsuji YAMAMOTO^1
1Graduate School of Engineering, Chiba University, Chiba, Chiba, Japan

The artifacts induced by magnetic susceptibility distribution cause serious problems for MR-guided therapeutic procedures. The effects of the devices on the perturbations need to be evaluated. However, the spatial resolution for field calculation is insufficient because of computer memory limitations. To overcome this challenge, we developed a computer code based on the superposition principle to increase the local resolution and on expansion of the calculation volume with extrapolation. We successfully applied the method to the needles penetrated into the realistic head model and demonstrated the susceptibility effects of the device. The computer code should be useful in designing MR-compatible devices.

Andreas Schäfer¹, Solveig Tiepolt², Elisabeth Roggenhofer¹, Robert Trampel¹, Carsten Stueber¹, Vilia Zeisig², Udo Grossmann², Thies H. Jochimsen², Osama Sabri², Turner Robert¹, and Henryk Barthel^2
1Max-Planck-Institute for Human Cognitive and Brain Sciences, Leipzig, Germany, ²Department of Nuclear Medicine, University of Leipzig, Germany

In this study we examined 6 patients with beta-amyloid-PET-confirmed Alzheimer disease and 10 healthy controls at 7T. Based on the phase data of a whole brain FLASH sequence with 0.7mm isotropic resolution we calculated the susceptibility maps. Across the group, QSM values in the AD patients were significantly increased compared to healthy controls. The regional analysis showed a significant increase of QSM values in regions typically affected in AD. This more paramagnetic behavior could be due to an increased iron accumulation in brain regions in Alzheimer disease, which may coincide already with early plaque formation.

Casey A Anderson¹, Kimberly R Pechman², and Kathleen M Schmainda^3,4
1Biophysics, Medical College of Wisconsin, Milwaukee, Wisconsin, United States, ²Neurosurgery, Medical College of Wisconsin, Milwaukee, WI, United States,³Radiology, Medical College of Wisconsin, Milwaukee, WI, United States, ⁴Biophysics, Medical College of Wisconsin, Milwaukee, WI, United States

With the increased understanding of iron metabolism and its relationship to cancer, iron-imaging techniques may provide valuable insight about the underlying physiology of tumors. While several imaging techniques can quantify iron, quantitative susceptibility mapping (QSM) is most dependent on the susceptibility effects from iron while the others are affected by tissue properties. We evaluated rat brain tumors with QSM to see if increased iron is found in the tumors. From the initial findings of this work, we demonstrate that increased susceptibility is detectable in rat brain tumors with QSM, and is a promising means to quantify iron in tumors.

Xue Xiao^1,2, Wei Li², Kui Ying³, and Chunlei Liu^2,4
1Department of Biomedical Engineering, Tsinghua University, Beijing, Beijing, China, ²Brain Imaging & Analysis Center, Duke University, Durham, NC, United States, ³Department of Engineering Physics, Tsinghua University, Beijing, China, ⁴Department of Radiology, Duke University, NC, United States

The analysis of white matter in Quantitative Susceptibility Mapping (QSM) is complicated as white matter is heterogeneous due to different fiber. To utilize QSM in population based imaging analysis, a normalized brain atlas of QSM is needed. We developed such an atlas for automatic analysis of magnetic susceptibility in white matter and proposed a method for white matter analysis of the whole brain. The atlas was employed to investigate magnetic susceptibility anisotropy according to a sine-squared relationship with the fiber orientation. Results indicated a generally negative relation between magnetic susceptibility and contribution of myelin in white matter, and the fiber-track-based atlas was found to be more useful for anisotropy analysis of the white matter compared to region-based atlas.

Xu Li^1,2, Hongjun Liu^1,3, Richard Allen⁴, Christopher J Earley⁴, Richard Edden^1,2, Peter B Barker^1,2, Tiana Krum⁴, and Peter C.M. van Zijl^1,2
1F.M. Kirby research center for functional brain imaging, Kennedy Krieger Institute, Baltimore, MD, United States, ²Radiology, Johns Hopkins University School of Medicine, Baltimore, MD, United States, ³Radiology, Guangdong General Hospital, Guangzhou, Guangdong, China, ⁴Neurology, Johns Hopkins University School of Medicine, Baltimore, MD, United States

Quantitative susceptibility mapping (QSM) at 7T was utilized to assess possible brain iron deficiency in restless legs syndrome (RLS) (AKA Willis Ekbom disease) using the measured tissue magnetic susceptibility as an iron index. Initial data collected on 14 RLS patients and age-matched normal controls (n=5) showed decreased magnetic susceptibility in RLS patients as compared to controls in substantia nigra, red nucleus, dentate nucleus, globus pallidus and the pulvinar nucleus of thalamus. Among them the difference in dentate nucleus was statistically significant (t test, one sided for directional hypotheses, p<0.05), while there was a trend toward differences in substantia nigra (p<0.12).

Xiaozhen Li^1,2, Peter van Gelderen¹, Pascal Sati³, Jacco A. de Zwart¹, Daniel S. Reich³, and Jeff H. Duyn^1
1Advanced MRI Section, LFMI, NINDS, National Institutes of Health, Bethesda, Maryland, United States, ²Division of Clinical Geriatrics, NVS Dept., Karolinska Institutet, Huddinge, Stockholm, Sweden, ³Translational Neuroradiology Unit, Neuroimmunology Branch, NINDS, National Institutes of Health, Bethesda, Maryland, United States

Multiple sclerosis (MS) is a chronic disease of the nervous system characterized by focal areas of myelin loss (“lesions”). Recent gradient-echo studies suggest the possibility of obtaining cellular compartment-specific information from multi-component fitting of the T2* relaxation decay curve, allowing determination of the relative fractions of myelin water, axonal water and interstitial water. Our findings suggested that three-component fitting of the T2* relaxation decay curve in MS lesions may help quantify cumulative myelin loss, and possibly discriminate between chronic and more acute stages of lesion evolution.

Kohsuke Kudo^1,2, Tian Liu³, Jonathan Goodwin², Ikuko Uwano², Fumio Yamashita², Satomi Higuchi², Noriyuki Fujima¹, Yi Wang³, Kuniaki Ogasawara⁴, Akira Ogawa⁴, and Makoto Sasaki^2
1Diagnostic and Interventional Radiology, Hokkaido University Hospital, Sapporo, Hokkaido, Japan, ²Ultra-High Field MRI, Iwate Medical University, Morioka, Iwate, Japan, ³Radiology, Weill Cornell Medical College, NY, United States, ⁴Neurosurgery, Iwate Medical University, Morioka, Iwate, Japan

Oxygen extraction fraction (OEF) represents an important parameter of brain metabolism. The purposes of this study are, to establish OEF measurements using quantitative susceptibility mapping (QSM), and to compare QSM-OEF with the gold standard PET-OEF in patients with unilateral chronic steno-occlusive disease. QSM-OEF value in the affected hemisphere was significantly higher than the contralateral hemisphere. Good correlation of OEF ratio between QSM-OEF and PET-OEF was observed, and the sensitivity and specificity for the detection of increased OEF by QSM was 0.80 and 0.90, respectively.

Ilhami Kovanlikaya¹, Apostolos John Tsiouris¹, Tian Liu², Jingwei Zhang³, Yi Wang^1,4, and Shixin Chang^5
1Radiology, Weill Cornell Medical College, New York, New York, United States, ²MedImageMetric LLC, New York, New York, United States, ³Biomedical Engineering, Cornell University, Ithaca, New York, United States, ⁴Biomedical Engineering, Cornell University, Itahca, New York, United States, ⁵Radiology, Yueyang Hospital of Integrated Traditional Chinese & Western Medicine, Shanghai, China

Identification and characterization of intracranial hemorrhages (ICH) are critical for appropriate management of hemorrhagic stroke patients. Hematomas at different stages have different appearance on T1 weighted, T2 weighted, and gradient echo images, partly because magnetic susceptibility varies over the time course. In this study, we applied quantitative susceptibility mapping (QSM) in vivo to directly investigate the magnetic susceptibility of hematomas at various stages.

Chunlei Liu^1,2
1Brain Imaging and Analysis Center, Duke University, Durham, NC, United States, ²Radiology, Duke University, Durham, NC, United States

Multipole anisotropy can be measured by STI in the Fourier spectral space. The Fourier spectral space, or p-space, can be generated by applying gradients or equivalently by shifting k-space data in various directions. Shifting k-pace, however, creates regions without data support that have to be zero filled. Concerns were raised that multipole anisotropy maybe an artifact of zero filling. Here, we demonstrated that this anisotropy is not caused by zero filling using mouse brain experiments. Our data support that p-space STI may enable practical mapping of tissue microstructure in vivo without rotating subject or magnetic field.

Huan Tan¹, Changbin Shi¹, Abdul Ghani Mikati¹, Robert Shenkar¹, Tian Liu², Yi Wang^3,4, Pottumarthi V Prasad^1,5, Robert R Edelman^5,6, and Issam Awad^1
1University of Chicago, Chicago, IL, United States, ²MedImageMetric LLC, New York, NY, United States, ³Weill Cornell Medical College, New York, NY, United States, ⁴Cornell University, Ithaca, NY, United States, ⁵NorthShore University HealthSystem, Evanston, IL, United States, ⁶Northwestern University Feinberg School of Medicine, Chicago, IL, United States

In this study, we validated quantitative susceptibility mapping (QSM) in two ex-vivo studies. First, we correlated QSM results in human cerebral cavernous malformation (CCM) lesion specimens against mass spectroscopy. Second, we compared lesional susceptibility distribution with histology in CCM mice. The preliminary results demonstrated the excellent sensitivity and specificity of QSM to identify and quantify lesional iron content in CCM.

Saurav ZK Sajib¹, Woo Chul Jeong¹, Hyung Joong Kim¹, Oh In Kwon², and Eung Je Woo^1
1Kyung Hee University, Yongin, Korea, ²Konkuk University, Seoul, Korea

DTI has been used as a powerful tool for investigating the anisotropic property of biological tissues. MR-based electrical impedance tomography (MREIT) could visualize the apparent tissue conductivity by measuring the one component of magnetic flux density by the externally injected currents. The diffusivity of water molecule and electrical tissue conductivity depend on the anisotropic orientation of tissue. The effective medium model implies that the diffusion and electrical conductivity tensor eigenvalues are connected through a linear relationship. Combing DTI and MREIT techniques, we propose a new method, DT-MREIT, which provides in vivo anisotropic absolute conductivity tensor images of the human body.

Jaewook Shin¹, Min-Oh Kim¹, Joonsung Lee¹, Narae Choi¹, and Dong-Hyun Kim^1
1Department of Electrical and Electronic Engineering, Yonsei University, seodaemungu, Seoul, Korea

Electrical conductivity of human brain tissue reveals tissue properties such as ion concentration, micro-structure and temperature. Due to these relationships between electrical conductivity value and biological factors, electrical conductivity has a potential to be a powerful bio-marker. Recently, in brain area, magnetic resonance electrical property tomography (MREPT) has been researched for clinical applications (systematic brain tumor) and biological characteristics (ion concentration and pH). However, due to lack of SNR and systemic error on tissue boundary, there were some restricts to quantify the electrical conductivity of sub-divisional brain tissue. Therefore, in this study, we try to observe conductivity values of sub-divisional brain tissue using weighted polynomial fitting technique with adaptively generated weighting factor. The weighting factor was generated using MPRAGE image that is well-optimized to enhance brain tissue contrast.

Oh In Kwon¹, Saurav ZK Sajib², Woo Chul Jeong², Hyung Joong Kim², and Eung Je Woo^2
1Konkuk University, Seoul, Korea, ²Kyung Hee University, Yongin, Korea

Magnetic resonance electrical impedance tomography (MREIT) is an emerging method to visualize static conductivity and/or current density images at low frequencies (below 1 kHz). Diffusion tensor imaging (DTI) measures the intrinsic three-dimensional diffusion property of water molecules within the human body. The effective macroscopic anisotropic tensor model implies the linear relationship between the effective electrical conductivity tensor and the water diffusion tensor. By assigning the linear relation, we propose a novel method to visualize the absolute anisotropic conductivity tensor map by combining DTI and MREIT techniques without any referred extracellular conductivity and diffusivity information.

Wenchao Cai¹, Feiyu Li¹, Jing Wang², Jue Zhang^3,4, Xiaoying Wang^1,3, and Xuexiang Jiang^1
1Peking University First Hospital, Beijing, Beijing, China, ²MRI Research Center, Academy for Advanced Interdisciplinary Studies, Peking University, Beijing, China, ³Academy for Advanced Interdisciplinary Studies, Peking University, Beijing, China, ⁴College of Engineering, Peking University, Beijing, China

Static blood oxygen level dependent (BOLD) MRI is a non-invasive method quantitatively measuring R2* which is sensitive to oxygenation status and accordance to hypoxia measured by immunohistochemical staining and pO2 microelectrode in prostate cancer. Tumor hypoxia have negative impact on response to various cancer therapies and promotion of metastasis. Therefore the purpose of the study was to investigate the ability of R2* by static BOLD MRI in staging of prostate tumor, and compared R2* with blood flow (BF) from ASL. Our results show a significant drop of R2* along with the prostate cancer upstaged detected via BOLD MRI, and inverse correlation between R2* and BF, which indicates that R2* values in patients with prostate cancer may be highly dependent on blood oxygenation.

Wei Li^1,2, Eric R. Muir¹, Bryan H. De La Garza¹, and Timothy Q Duong^1,2
1Research Imaging Institute, University of Texas Health Science Center at San Antonio, San Antonio, TX, United States, ²Ophthalmology, University of Texas Health Science Center at San Antonio, San Antonio, TX, United States

This study explored the use of high-resolution quantitative susceptibility mapping-based MRI microscopy for studying the highly stratified layers ex vivo rat retina at a spatial resolution of 23 x 23 x 57 m³. Comparing to magnitude with weak contrast, both gradient echo phase and magnetic susceptibility shows remarkable contrast between different layers. These layered structures from phase and susceptibility is in good agreement with H&E staining using histology.

Fei Cong¹, Bo Wang¹, Xiaohong Joe Zhou², Yan Zhuo¹, and Yongquan Ye^3
1Institute of Biophysics, Chinese Academy of Sciences, Chaoyang District, Beijing, China, ²Department of Radiology and Center for MR Research, University of Illinois Medical Center, Chicago, IL, United States, ³Department of Radiology, School of Medicine, Wayne State University, Detroit, MI, United States

We introduced a new single echo MRAV method with inversed vessel-tissue contrast, which shows arteries with dark blood contrast via low VENC flow dephasing and veins with bright blood contrast using quantitative susceptibility mapping. This method has the potential to reliably extract both arterial and venous networks separately from the data from single echo GRE data, without the penalty of total scan time or image misregistration. The reconstructed arterial and venous maps are shown, and the potential application and future works of this method are discussed.

Sergey Cheshkov^1,2, Guillaume Gilbert³, Ivan Dimitrov^1,4, Samantha By⁵, Joseph Rispoli⁵, Mary McDougall^5,6, Steve Wright^5,6, Stephen Seiler², and Craig Malloy^1,2
1Advanced Imaging Research Center, University of Texas Southwestern Medical Center, Dallas, TX, United States, ²Radiology, University of Texas Southwestern Medical Center, Dallas, TX, United States, ³Philips Medical Systems, Montréal, Québec, Canada, ⁴Philips Medical Systems, Cleaveland, OH, United States, ⁵Biomedical Engineering, Texas A&M University, College Station, TX, United States, ⁶Electrical Engineering, Texas A&M University, College Station, TX, United States

Pre-invasive cancer – ductal carcinoma in situ (DCIS) may frequently be missed by the clinical dynamic contrast-enhanced MRI. DCIS has been shown to strongly associate with particular microcalcifications distributions and patterns. Susceptibility weighted imaging (SWI) has been proposed as a non-invasive radiation-free alternative to mammography for early detection of DCIS-associated breast microcalcifications (diamagnetic), providing better localization and 3D morphology. However, the desired high spatial resolution is not feasible on the 1.5T and 3T systems due to the long acquisition times. Benefiting from higher susceptibility effects and gain in SNR we demonstrate ultra-high resolution 7T SWI in vivo for calcification detection.

Takoua Kaaouana^1,2, Ludovic de Rochefort³, Thomas Samaille^1,2, Nathalie Thierry⁴, Genevieve Chene⁴, Christine Delmaire⁵, Didier Dormont⁶, and Marie Chupin^1
1CRICM, UPMC UMR_S975, INSERM U975, CNRS UMR7225, ICM, INRIA, Paris, France, ²CATI, Paris, France, ³Univ. Paris-Sud, CNRS IR4M UMR8081, Orsay, France, ⁴U897, INSERM, Bordeaux, France, ⁵Neuroradiology, CHRU Roger Salengro, Lille, France, ⁶Neuroradiology, CHRU pitie salpetriere, Paris, France

In MRI signal, phase is proportional to the local resonance frequency and therefore directly reflects magnetic field changes induced by local magnetic susceptibility sources. Composed of hemosiderin, cerebral microbleed can act as susceptibility sources and generate local dipole magnetic field. Here, we introduce a physically motivated fast 2D phase processing technique including unwrapping and harmonic filtering which yields the internal field. We demonstrate its applicability and robustness on multicenter 2D datasets acquired in standardized clinical setting to discriminate cerebral microbleeds.

Alexander Ciritsis¹, Christoph Wilkmann¹, Daniel Truhn¹, Christiane K. Kuhl¹, and Nils A. Krämer^1
1Department of Diagnostic and Interventional Radiology, RWTH University Hospital Aachen, Aachen, NRW, Germany

The purpose of this study was to evaluate a method analysing the mesh related susceptibility gradients to assess mesh shrinkage. Three iron oxide-loaded MR-visible mesh implants in different configurations (shrunk from 14 to 48 % )were placed in an agarose phantom. MRI was performed at a 1.5 T scanner using a GRE sequences and SG- maps were calculated. A Fourier analysis was performed on ten small rectangular areas, resulting in one frequency spectrum indicating the alternating susceptibility induced magnetic field gradients. We propose an approach to assess local configuration changes in iron oxide- loaded mesh implants without extensive 3D postprocessing.

Daniel Truhn¹, Alexander Ciritsis², Nienke Hansen², Alexandra Barabasch², Jens Otto², Christiane Kuhl², and Nils Kraemer^2
1University Hospital Aachen, Aachen, NRW, Germany, ²University Hospital Aachen, NRW, Germany

We did a comparison of positive contrast techniques (namely positive contrast susceptibility imaging and susceptibility gradient maps) in the visualization of iron-loaded mesh implants in humans. As a result we found that SGM offers better visualization of the mesh compared to PCSI.

Shan Hu¹, Wen zhen Zhu¹, IIhami Kovanlikaya², Kofi Deh², Tian Liu², and Yi Wang^2
1Department of Radiology, Tongji hospital. Tongji Medical College, WuHan, Hu Bei, China, ²Department of Radiology, Weill Cornell Medical College, New York, United States

Quantitative Susceptibility Mapping (QSM) a new quantitative imaging technique may directly demonstrate the dia and paramagnetic properties of intra-tumor. We investigated whether the new post processing approach is able to diagnose and grade meningiomas.

Wei Li^1,2, Christian Langkammer³, Katja Petrovic³, Reinhold Schmidt³, Allen W Song^1,4, Stefan Ropele³, and Chunlei Liu^1,4
1Brain Imaging and Analysis Center, Duke University, Durham, NC, United States, ²Research Imaging Institute, University of Texas Health Science Center at San Antonio, San Antonio, TX, United States, ³Neurology, Medical University of Graz, Graz, Austria, ⁴Radiology, Duke University, Durham, NC, United States

In human brain, certain deep gray matter structures, particularly globus pallidus, putamen, caudate nucleus, substantia nigra, red nucleus, and dentate nucleus have especially high iron contents compared to other brain tissues. In this study, we correlated the magnetic susceptibility – a surrogate marker of brain iron – of these gray matter structures with several clinical measures of brain function in 135 healthy adults aged 40~83 years. We observed significant negative correlations between the scores of motor function, language skills, transit memory, and executive functions with the magnetic susceptibility of one or several of the aforementioned deep brain structures.

Pei-Shan Ho¹, Ting-Yu Chang², Kuo-Lun Huang², Feng-Xian Yan^1,3, Ho-Fai Wong⁴, Tsong-Hai Lee², and Ho-Ling Liu^1,4
1Medical Imaging and Radiological Sciences, Chang Gung University, Taoyuan, Taiwan, ²Neurology and Stroke Center, Chang Gung Memorial Hospital and Chang Gung University College of Medicine, Taoyuan, Taiwan, ³Radiology, Taipei Medical University - Shuang Ho Hospital, New Taipei City, Taiwan, ⁴Medical Imaging and Intervention, Chang Gung Memorial Hospital, Taoyuan, Taiwan

Severe internal carotid artery (ICA) stenosis often accompanies reversed ophthalmic artery (OA) flow; however, the quantified comparison between OA flow forward and reversed was not documented. This study compared the indicative parameter maps of dynamic susceptibility contrast (DSC) MRI between subjects suffering from unilateral ICA stenosis with and without OA flow reversed. 51 unilateral ICA stenosis patients underwent DSC MRI before and after stenting. Contrary to patients with OA flow forward, patients with OA flow reversed have increased mean transit time as well as arrival time delay before stent surgery. Moreover, the effectively recovery was detectable after carotid stent replacement.

Ying Dong¹, Zheng Chang², Gregory Whitehead³, and Jim Xiuquan Ji^4
1Electrical Engineering, Texas A&M University, College Station, Texas, United States, ²Radiation Oncology, Duke University, NC, United States, ³Texas A&M University, College Station, TX, United States, ⁴Texas A&M University, College Station, Texas, United States

For a number of reasons, MRI is preferred for brachytherapy imaging. However, due to the high susceptibility of the seeds and the lack of protons, brachytherapy seeds and the surrounding tissues usually show as enlarged dark spots (negative contrast) on the MR images. Many methods have been proposed to create positive contrast in the surrounding tissues (but not in the seeds) based on the susceptibility-induced field change. This paper proposes a new method to image the brachytherapy seeds by deriving the susceptibility distribution. The method is based on an improved, sparsity-promoting, kernel deconvolution algorithm. Experimental results show the proposed method can provide localization of the seed itself, rather than the surrounding tissues, and it gives an effective positive contrast to the seeds. In addition, the proposed method is also able to identify and differentiate the seeds from other less-susceptible objects.

James Goldfarb^1,2 and Usama Hasan^1
1Department of Research and Education, St Francis Hospital, Roslyn, NY, United States, ²Program in Biomedical Engineering, Stony Brook University, Stony Brook, NY, United States

We compared single-echo and a novel multiple-echo image combination using source or susceptibility weighted images (SWI) for improvements in image quality assessed quantitatively through intramyocardial hemorrhage contrast and signal-difference-to-noise measurements. SWI with echo-combination was an effective means of generating high quality images of left ventricular myocardium for tissue characterization. Echo-combination of source or SWI images increases SNR at the cost of hemorrhage contrast. SWI with six phase mask multiplications and 10 echo averages provides optimal image SDNR and contrast. Image quality of echo-combined SWI is superior to source, SWI and echo-combined source images.