Faculty Bibliography

INTRODUCTION/UNASSIGNED:Femoroacetabular Impingement (FAI) is a hip pathology characterized by impingement of the femoral head-neck junction against the acetabular rim, due to abnormalities in bone morphology. FAI is normally diagnosed by manual evaluation of morphologic features on magnetic resonance imaging (MRI). In this study, we assess, for the first time, the feasibility of using radiomics to detect FAI by automatically extracting quantitative features from images. MATERIAL AND METHODS/UNASSIGNED:-value, for a total of 48 subsets. For each subset, a K-nearest neighbor model was trained to differentiate between IJ and HJ using the values of the radiomic features in the subset as input. The training was repeated 100 times, randomly subdividing the data with 75%/25% training/testing. RESULTS/UNASSIGNED:The texture-based gray level features yielded the highest prediction max accuracy (0.972) with the smallest subset of features. This suggests that the gray image values are more homogeneously distributed in the HJ in comparison to IJ, which could be due to stress-related inflammation resulting from impingement. CONCLUSIONS/UNASSIGNED:We showed that radiomics can automatically distinguish IJ from HJ using water-only Dixon MRI. To our knowledge, this is the first application of radiomics for FAI diagnosis. We reported an accuracy greater than 97%, which is higher than the 90% accuracy for detecting FAI reported for standard diagnostic tests (90%). Our proposed radiomic analysis could be combined with methods for automated joint segmentation to rapidly identify patients with FAI, avoiding time-consuming radiological measurements of bone morphology.

We propose a framework to interpret the effects of High Permittivity Materials (HPM) on the performance of radiofrequency coils in Magnetic Resonance Imaging (MRI). Based on a recent formulation of the scattering and propagation properties of spheres, we expanded the field in a layered sphere as a superposition of ingoing and outgoing travelling waves, which allowed us to study the field components with a non-homogeneous transmission line model. We investigated the effects of a layer of HPM surrounding a head-mimicking uniform sphere at 7 tesla. Through the analysis of impedance and reflection coefficients, we show that by adjusting the properties of the HPM, it is possible to selectively amplify individual modes, or combination of them, modifying the overall field distribution in the sample and increasing signal-to-noise ratio at specific locations. Our results demonstrate that the observed enhanced MRI performance is not merely due to secondary fields generated by displacement currents in the HPM. Our formulation explains the effect of HPM in terms of matching, enabling the optimization of the electrical properties of the HPM with a simple mode-matching formula. The proposed method could guide the design of novel radiofrequency coils with integrated HPM.

We investigated how to construct low-order subspace basis sets to accurately represent electromagnetic fields generated within inhomogeneous arbitrary objects by radio-frequency sources external to a Huygen's surface. The basis generation relies on the singular value decomposition of Green's functions integro-differential operators which makes it feasible to derive a reduced-order yet stable model. We present a detailed study of the theoretical and numerical requisites for generating such basis, and show how it can be used to calculate performance limits in magnetic resonance imaging applications. Finally, we propose a novel numerical framework for the computation of characteristic modes of arbitrary inhomogeneous objects. We validated accuracy and convergence properties of the numerical basis against a complete analytical basis in the case of a uniform spherical object. We showed that the discretization of the Huygens's surface has a minimal effect on the accuracy of the calculations, which mainly depended on the electromagnetic solver resolution and order of approximation.

PURPOSE/OBJECTIVE:To present a novel 3T 24-channel glove array that enables hand and wrist imaging in varying postures. METHODS:The glove array consists of an inner glove holding the electronics and an outer glove protecting the components. The inner glove consists of four main structures: palm, fingers, wrist, and a flap that rolls over on top. Each structure was constructed out of three layers: a layer of electrostatic discharge flame-resistant fabric, a layer of scuba neoprene, and a layer of mesh fabric. Lightweight and flexible high impedance coil (HIC) elements were inserted into dedicated tubes sewn into the fabric. Coil elements were deliberately shortened to minimize the matching interface. Siemens Tim 4G technology was used to connect all 24 HIC elements to the scanner with only one plug. RESULTS:The 24-channel glove array allows large motion of both wrist and hand while maintaining the SNR needed for high-resolution imaging. CONCLUSION/CONCLUSIONS:In this work, a purpose-built 3T glove array that embeds 24 HIC elements is demonstrated for both hand and wrist imaging. The 24-channel glove array allows a great range of motion of both the wrist and hand while maintaining a high SNR and providing good theoretical acceleration performance, thus enabling hand and wrist imaging at different postures to extract kinematic information.

We investigated how to construct low-order subspace basis sets to accurately represent electromagnetic fields generated within inhomogeneous arbitrary objects by radio-frequency sources external to a Huygen’s surface. The basis generation relies on the singular value decomposition of Green’s functions integro-differential operators which makes it feasible to derive a reduced-order yet stable model. We present a detailed study of the theoretical and numerical requisites for generating such basis, and show how it can be used to calculate performance limits in magnetic resonance imaging applications. Finally, we propose a novel numerical framework for the computation of characteristic modes of arbitrary inhomogeneous objects. We validated accuracy and convergence properties of the numerical basis against a complete analytical basis in the case of a uniform spherical object. We showed that the discretization of the Huygens’s surface has a minimal effect on the accuracy of the calculations, which mainly depended on the electromagnetic solver resolution and order of approximation.

In this work, we propose a method for the compression of the coupling matrix in volume-surface integral equation (VSIE) formulations. VSIE methods are used for electromagnetic analysis in magnetic resonance imaging (MRI) applications, for which the coupling matrix models the interactions between the coil and the body. We showed that these effects can be represented as independent interactions between remote elements in 3D tensor formats, and subsequently decomposed with the Tucker model. Our method can work in tandem with the adaptive cross approximation technique to provide fast solutions of VSIE problems. We demonstrated that our compression approaches can enable the use of VSIE matrices of prohibitive memory requirements, by allowing the effective use of modern graphical processing units (GPUs) to accelerate the arising matrix-vector products. This is critical to enable numerical MRI simulations at clinical voxel resolutions in a feasible computation time. In this paper, we demonstrate that the VSIE matrix-vector products needed to calculate the electromagnetic field produced by an MRI coil inside a numerical body model with 1 mm³ voxel resolution, could be performed in ~ 33 seconds in a GPU, after compressing the associated coupling matrix from ~ 80 TB to ~ 43 MB.

BACKGROUND:Signal-to-noise ratio (SNR) is used to evaluate the performance of magnetic resonance (MR) imaging systems. Accurate and consistent estimations are needed to routinely use SNR to assess coils and image reconstruction techniques. PURPOSE/OBJECTIVE:To identify a reliable and practical method for SNR estimation in multiple-coil reconstructions. STUDY TYPE/METHODS:Technical evaluation and comparison. SUBJECTS/PHANTOM/UNASSIGNED:A uniform phantom and four healthy volunteers: 35, 38, 39 y/o males, 25 y/o female. FIELD STRENGTH/SEQUENCE/UNASSIGNED:Two-dimensional multislice gradient-echo pulse sequence at 3 T and 7 T. ASSESSMENT/RESULTS:Reference-standard SNR was calculated from 100 multiple replicas. Six SNR methods were compared against it: difference image (DI), analytic array combination (AC), pseudo-multiple-replica (PMR), generalized pseudo-replica (GPR), smoothed image subtraction (SIS), and DI with temporal instability correction (TIC). The assessment was repeated for different multiple-coil reconstructions. STATISTICAL TESTS/UNASSIGNED:SNR methods were evaluated in terms of relative deviation (RD) and normalized mutual information (NMI) with respect to the reference-standard, using a linear regression (0.05 significance level) to assess how different factors affect accuracy. RESULTS:Average RD (phantom) for DI, AC, PMR, GPR, SIS, and TIC was 7.9%, 6%, 6.7%, 10.1%, 40%, and 14.6%, respectively. RD increased with acceleration. SNR maps with AC were the most similar to the reference standard (NMI = 0.358). Considering all brain regions of interest, average RD for all SNR methods varied 96% among volunteers but remained approximately 10% for AC, PMR, and GPR, whereas it was more than 30% for DI, SIS, and TIC. RD was mainly affected by image reconstruction (beta = 12) for AC and SNR entropy for SIS (beta = 19). DATA CONCLUSION/UNASSIGNED:AC provided accurate and robust SNR estimation. PMR and GPR are more generally applicable than AC. DI and TIC should be used only at low acceleration factors, when an additional noise-only scan cannot be acquired. SIS is a single-acquisition alternative to DI for generalized autocalibrating partial parallel acquisition (GRAPPA) reconstructions. EVIDENCE LEVEL/UNASSIGNED:1 TECHNICAL EFFICACY: Stage 1.

PURPOSE/OBJECTIVE:Investigating the designs and effects of high dielectric constant (HDC) materials in the shape of a conformal helmet on the enhancement of RF field and reduction of specific absorption rate at 10.5 T for human brain studies. METHODS:A continuous and a segmented four-piece HDC helmet fit to a human head inside an eight-channel fractionated-dipole array were constructed and studied with a phantom and a human head model using computer electromagnetic simulations. The simulated transmit efficiency and receive sensitivity were experimentally validated using a phantom with identical electric properties and helmet-coil configurations of the computer model. The temporal and spatial distributions of displacement currents on the HDC helmets were analyzed. RESULTS:Using the continuous HDC helmet, simulation results in the human head model demonstrated an average transmit efficiency enhancement of 66%. A propagating displacement current was induced on the continuous helmet, leading to an inhomogeneous RF field enhancement in the brain. Using the segmented four-piece helmet design to reduce this effect, an average 55% and 57% enhancement in the transmit efficiency and SNR was achieved in human head, respectively, along with 8% and 28% reductions in average and maximum local specific absorption rate. CONCLUSION/CONCLUSIONS:The HDC helmets enhanced the transmit efficiency and SNR of the dipole array coil in the human head at 10.5 T. The segmentation of the helmet to disrupt the continuity of circumscribing displacement currents in the helmet produced a more uniform distribution of the transmit field and lower specific absorption rate in the human head compared with the continuous helmet design.