arXiv subjects

Xiaoliang Zhang

Publications and source records attributed to Xiaoliang Zhang.

33 records · Page 2Linked to original sources

arXiv2023

Self-supervised Deep Unrolled Reconstruction Using Regularization by Denoising

Deep learning methods have been successfully used in various computer vision tasks. Inspired by that success, deep learning has been explored in magnetic resonance imaging (MRI) reconstruction. In particular, integrating deep learning and model-based optimization methods has shown considerable advantages. However, a large amount of labeled training data is typically needed for high reconstruction quality, which is challenging for some MRI applications. In this paper, we propose a novel reconstruction method, named DURED-Net, that enables interpretable self-supervised learning for MR image reconstruction by combining a self-supervised denoising network and a plug-and-play method. We aim to boost the reconstruction performance of Noise2Noise in MR reconstruction by adding an explicit prior that utilizes imaging physics. Specifically, the leverage of a denoising network for MRI reconstruction is achieved using Regularization by Denoising (RED). Experiment results demonstrate that the proposed method requires a reduced amount of training data to achieve high reconstruction quality among the state-of-art of MR reconstruction utilizing the Noise2Noise method.

eess.IV↗

arXiv2022

Double Cross Magnetic Wall Decoupling for Quadrature Transceiver RF Array Coils using Common-Mode Differential-mode Resonators

In contrast to linearly polarized RF coil arrays, quadrature transceiver coil arrays are capable of improving signal-to-noise ratio (SNR), spatial resolution and parallel imaging performance. Owing to a reduced excitation power, low specific absorption rate can be also obtained using quadrature RF coils. However, due to the complex nature of their structure and their electromagnetic proprieties, it is challenging to achieve sufficient electromagnetic decoupling while designing multichannel quadrature RF coil arrays, particularly at ultrahigh fields. In this work, we proposed a double cross magnetic wall decoupling for quadrature transceiver RF arrays and implemented the decoupling method on common-mode differential mode quadrature (CMDM) quadrature transceiver arrays at ultrahigh field of 7T. The proposed magnetic decoupling wall comprised of two intrinsic decoupled loops is used to reduce the mutual coupling between all the multi-mode current present in the quadrature CMDM array. The decoupling network has no physical connection with the CMDMs' coils giving leverage over size adjustable RF arrays. In order to validate the feasibility of the proposed cross magnetic decoupling wall, systematic studies on the decoupling performance based on the impedance of two intrinsic loops are numerically performed. A pair of quadrature transceiver CMDMs is constructed along with the proposed decoupling network and their scattering matrix is characterized using a network analyzer. The measured results show all the current modes coupling are concurrently suppressed using the proposed cross magnetic wall. Moreover, field distribution, and SNR intensity are numerically obtained for a well-decoupled 8-channel quadrature knee-coil array.

physics.med-ph↗

arXiv2022

Visual Area of Interests based Multimodal Trajectory Prediction for Probabilistic Risk Assessment

Accurate and reliable prediction of driving intentions and future trajectories contributes to cooperation between human drivers and ADAS in complex traffic environments. This paper proposes a visual AOI (Area of Interest) based multimodal trajectory prediction model for probabilistic risk assessment at intersections. In this study, we find that the visual AOI implies the driving intention and is about 0.6-2.1 s ahead of the operation. Therefore, we designed a trajectory prediction model that integrates the driving intention (DI) and the multimodal trajectory (MT) predictions. The DI model was pre-trained independently to extract the driving intention using features including the visual AOI, historical vehicle states, and environmental context. The intention prediction experiments verify that the visual AOI-based DI model predicts steering intention 0.925 s ahead of the actual steering operation. The trained DI model is then integrated into the trajectory prediction model to filter multimodal trajectories. The trajectory prediction experiments show that the proposed model outperforms the state-of-the-art models. Risk assessment for traffics at intersections verifies that the proposed method achieves high accuracy and a low false alarm rate, and identifies the potential risk about 3 s before a conflict occurs.

cs.HC↗

arXiv2022

A Multichannel RF Transceiver Array with mixed L-C loop and microstrip elements for Foot/Ankle MR Imaging at 7T

It is technically challenging to design efficient transceiver coil arrays for foot and ankle imaging at ultrahigh fields due to the irregular geometry of the anatomy. Shortened wavelength because of the high operation frequency at ultrahigh fields increases phase variation, leading to inhomogeneous B1 distribution in the imaging sample. In this work, we propose a hybrid design with mixed L-C loops and microstrip resonators for multichannel foot/ankle transceiver arrays at the ultrahigh field of 7T. The proposed transceiver array consists of 14 microstrip resonators and 5 L-C loop coils to cover the entire region of interest of the irregular-shaped foot/ankle with a relatively uniform B1 distribution at 7T. The feasibility and field behavior of the proposed design are systematically investigated numerically.

physics.med-ph↗

arXiv2022

Hairpin RF Resonators for Transceiver Arrays with High Inter-channel Isolation and B1 Efficiency at Ultrahigh Field 7T MR Imaging

Electromagnetic decoupling among a close-fitting or high-density transceiver RF array elements is required to maintain the integrity of the magnetic flux density from individual channel for enhanced performance in detection sensitivity and parallel imaging. High-impedance RF coils have demonstrated to be a prominent design method to circumvent these coupling issues. Yet, inherent characteristics of these coils have ramification on the B1 field efficiency and SNR. In this work, we propose a hairpin high impedance RF resonator design for highly decoupled multichannel transceiver arrays at ultrahigh magnetic fields. Due to the high impedance property of the hairpin resonators, the proposed transceiver array can provide high decoupling performance without using any dedicated decoupling circuit among the resonant elements. Because of elimination of lumped inductors in the resonator circuit, higher B1 field efficiency in imaging subjects can be expected. In order to validate the feasibility of the proposed hairpin RF coils, systematical studies on decoupling performance, field distribution, and SNR are performed, and the results are compared with those obtained from existing high-impedance RF coil, e.g., "self-decoupled RF coil". To further investigate its performance, an 8-channel head coil array using the proposed hairpin resonators loaded with a cylindrical phantom is designed, demonstrating a 19 % increase of the B1+ field intensity compared to the "self-decoupled" coils at 7T. Furthermore, the characteristics of the hairpin RF coils are evaluated using a more realistic human head voxel model numerically. The proposed hairpin RF coil provides excellent decoupling performance and superior RF magnetic field efficiency compared to the self-decoupled high impedance coils. Bench test of a pair of fabricated hairpin coils prove to be in good accordance with numerical results.

physics.med-ph↗

arXiv2022

High Dielectric Sheet to reduce Electric Fields and Flatten Magnetic Fields in Self-decoupled Radiofrequency coils for MR Imaging

High impedance RF coils, such as self-decoupled coils, reduce the electromagnetic coupling between the coil elements and eliminates the use of complex decoupling technologies. Although the high impedance design promises excellent decoupling between the coil elements, it also results in high electric fields across the RF coil, leading to potential safety problems during imaging. It also causes B1 field asymmetry, ultimately leading to difficulties in imaging quantification. In this study, we propose and investigate using a high dielectric sheet to reduce the electric fields across the coil and achieve excellent electromagnetic decoupling among the coil elements, thereby ensuring safer MRI at ultra-high fields and maintaining high imaging performance.

physics.med-ph↗

arXiv2022

Design of a 13-Channel Hybrid Array System for Foot/Ankle Magnetic Resonance Imaging at 7T/300MHz

Microstrip lines are being used in MR applications due to their unique properties, such as reduced radiation loss, high-frequency capability, and reduced perturbation of sample loading to the RF coil compared to conventional coils. Here, we present the design of the 13-channel hybrid array consisting of 12 Microstrips, 1 volume half birdcage coil placed on the foot/ankle phantom, and high permittivity materials to cover the maximum area of the subject at 7T/300MHz. We demonstrate using electromagnetic simulations, magnetic field distribution, SAR performance, and the coupling performance of the array elements. This work provides an ultrahigh field multichannel RF solution to lower extremity MR imaging with excellent imaging coverage and field uniformity.

physics.med-ph↗

arXiv2022

Accessing negative Poisson`s ratio of graphene by machine learning interatomic potentials

The negative Poisson`s ratio (NPR) is a novel property of materials, which enhances the mechanical feature and creates a wide range of application prospects in lots of fields, such as aerospace, electronics, medicine, etc. Fundamental understanding on the mechanism underlying NPR plays an important role in designing advanced mechanical functional materials. However, with different methods used, the origin of NPR is found different and conflicting with each other, for instance, in the representative graphene. In this study, based on machine learning technique, we constructed a moment tensor potential (MTP) for molecular dynamics (MD) simulations of graphene. By analyzing the evolution of key geometries, the increase of bond angle is found to be responsible for the NPR of graphene instead of bond length. The results on the origin of NPR are well consistent with the start-of-art first-principles, which amend the results from MD simulations using classic empirical potentials. Our study facilitates the understanding on the origin of NPR of graphene and paves the way to improve the accuracy of MD simulations being comparable to first-principle calculations. Our study would also promote the applications of machine learning interatomic potentials in multiscale simulations of functional materials. *Author

physics.comp-ph↗

arXiv2021

A Compact Planar Triple-Nuclear Coil for Small Animal 1H, 13C, and 31P Metabolic MR Imaging at 14.1 T

Multi-nuclear radio-frequency (RF) coils at ultrahigh field strengths are challenging to develop due to the high operating frequency, increased electromagnetic interaction among the coil elements, and increased electromagnetic interaction between the coils and the subject. In this work, a triple-nuclear surface coil for 1H, 13C, and 31P for small animal metabolic imaging at 14.1 T is built by nesting three surface coil geometries: a lumped-element L-C loop coil, a butterfly coil, and a microstrip transmission line (MTL) resonator. The loop coil, butterfly coil, and MTL are tuned to 31P, 13C, and 1H, respectively. The successful bench tests and phantom imaging with this novel triple-nuclear coil demonstrates the feasibility of the design for triple-nuclear MR imaging and spectroscopy studies at the ultrahigh field of 14.1 T. In particular, the microstrip transmission line resonator demonstrates excellent coil performance for proton imaging at 600 MHz, where conventional lumped-element design is limited.

physics.med-ph↗

arXiv2021

SuperDTI: Ultrafast diffusion tensor imaging and fiber tractography with deep learning

Purpose: To propose a deep learning-based reconstruction framework for ultrafast and robust diffusion tensor imaging and fiber tractography. Methods: We propose SuperDTI to learn the nonlinear relationship between diffusion-weighted images (DWIs) and the corresponding tensor-derived quantitative maps as well as the fiber tractography. Super DTI bypasses the tensor fitting procedure, which is well known to be highly susceptible to noise and motion in DWIs. The network is trained and tested using datasets from Human Connectome Project and patients with ischemic stroke. SuperDTI is compared against the state-of-the-art methods for diffusion map reconstruction and fiber tracking. Results: Using training and testing data both from the same protocol and scanner, SuperDTI is shown to generate fractional anisotropy and mean diffusivity maps, as well as fiber tractography, from as few as six raw DWIs. The method achieves a quantification error of less than 5% in all regions of interest in white matter and gray matter structures. We also demonstrate that the trained neural network is robust to noise and motion in the testing data, and the network trained using healthy volunteer data can be directly applied to stroke patient data without compromising the lesion detectability. Conclusion: This paper demonstrates the feasibility of superfast diffusion tensor imaging and fiber tractography using deep learning with as few as six DWIs directly, bypassing tensor fitting. Such a significant reduction in scan time may allow the inclusion of DTI into the clinical routine for many potential applications.

eess.IV↗

arXiv2020

Decoupled Self Attention for Accurate One Stage Object Detection

As the scale of object detection dataset is smaller than that of image recognition dataset ImageNet, transfer learning has become a basic training method for deep learning object detection models, which will pretrain the backbone network of object detection model on ImageNet dataset to extract features for classification and localization subtasks. However, the classification task focuses on the salient region features of object, while the location task focuses on the edge features of object, so there is certain deviation between the features extracted by pretrained backbone network and the features used for localization task. In order to solve this problem, a decoupled self attention(DSA) module is proposed for one stage object detection models in this paper. DSA includes two decoupled self-attention branches, so it can extract appropriate features for different tasks. It is located between FPN and head networks of subtasks, so it is used to extract global features based on FPN fused features for different tasks independently. Although the network of DSA module is simple, but it can effectively improve the performance of object detection, also it can be easily embedded in many detection models. Our experiments are based on the representative one-stage detection model RetinaNet. In COCO dataset, when ResNet50 and ResNet101 are used as backbone networks, the detection performances can be increased by 0.4% AP and 0.5% AP respectively. When DSA module and object confidence task are applied in RetinaNet together, the detection performances based on ResNet50 and ResNet101 can be increased by 1.0% AP and 1.4% AP respectively. The experiment results show the effectiveness of DSA module. Code is at: https://github.com/chenzuge1/DSANet.git.

cs.CV↗

arXiv2020

An open boundary condition for high-order solutions of magnetohydrodynamics on unstructured grids

In this paper a characteristics-based open boundary condition (CBC) is proposed for the magnetohydrodynamic (MHD) system of equations. The algorithm is carefully designed and implemented in the context of a high-order flux reconstruction (FR) scheme under the Generalized Lagrange Multiplier (GLM)-MHD system of equations. It is implemented by adding the contribution of the characteristic equation directly to the corrected flux term in the FR scheme dispensing with solving time-dependent characteristic equations along boundary faces. The CBC method is shown to be more accurate and robust than commonly used zero normal derivative (ZND) and approximate Riemann solver boundary conditions (ARBC) in solving 1D, 2D, and 3D test problems. The CBC method is successfully applied to simulate challenging problems of magnetic reconnection for which other options failed to get stable results over long-period time integration.

physics.comp-ph↗

arXiv2018

A Random Sample Partition Data Model for Big Data Analysis

Big data sets must be carefully partitioned into statistically similar data subsets that can be used as representative samples for big data analysis tasks. In this paper, we propose the random sample partition (RSP) data model to represent a big data set as a set of non-overlapping data subsets, called RSP data blocks, where each RSP data block has a probability distribution similar to the whole big data set. Under this data model, efficient block level sampling is used to randomly select RSP data blocks, replacing expensive record level sampling to select sample data from a big distributed data set on a computing cluster. We show how RSP data blocks can be employed to estimate statistics of a big data set and build models which are equivalent to those built from the whole big data set. In this approach, analysis of a big data set becomes analysis of few RSP data blocks which have been generated in advance on the computing cluster. Therefore, the new method for data analysis based on RSP data blocks is scalable to big data.

cs.DC↗

arXiv2016

Methodology for determining the electronic thermal conductivity of metals via direct non-equilibrium ab initio molecular dynamics

Many physical properties of metals can be understood in terms of the free electron model, as proven by the Wiedemann-Franz law. According to this model, electronic thermal conductivity ($κ_{el}$) can be inferred from the Boltzmann transport equation (BTE). However, the BTE does not perform well for some complex metals, such as Cu. Moreover, the BTE cannot clearly describe the origin of the thermal energy carried by electrons or how this energy is transported in metals. The charge distribution of conduction electrons in metals is known to reflect the electrostatic potential (EP) of the ion cores. Based on this premise, we develop a new methodology for evaluating $κ_{el}$ by combining the free electron model and non-equilibrium ab initio molecular dynamics (NEAIMD) simulations. We demonstrate that the kinetic energy of thermally excited electrons originates from the energy of the spatial electrostatic potential oscillation (EPO), which is induced by the thermal motion of ion cores. This method directly predicts the $κ_{el}$ of pure metals with a high degree of accuracy.

cond-mat.mtrl-sci↗

arXiv2015

Quantitatively Analyzing Phonon Spectral Contribution of Thermal Conductivity Based on Non-Equilibrium Molecular Dynamics Simulation I: From Space Fourier Transform

Probing detailed spectral dependence of phonon transport properties in bulk materials is critical to improve the function and performance of structures and devices in a diverse spectrum of technologies. Currently, such information can only be provided by the phonon spectral energy density (SED) or equivalently time domain normal mode analysis (TDNMA) methods in the framework of equilibrium molecular dynamics simulation (EMD), but has not been realized in non-equilibrium molecular dynamics simulations (NEMD) so far. In this paper we generate a new scheme directly based on NEMD and lattice dynamics theory, called time domain direct decomposition method (TDDDM), to predict the phonon mode specific thermal conductivity. Two benchmark cases of Lennard-Jones (LJ) Argon and Stillinger-Weber (SW) Si are studied by TDDDM to characterize contributions of individual phonon modes to overall thermal conductivity and the results are compared with that predicted using SED and TDNMA. Excellent agreements are found for both cases, which confirm the validity of our TDDDM approach. The biggest advantage of TDDDM is that it can be used to investigate the size effect of individual phonon modes in NEMD simulations, which cannot be tackled by SED and TDNMA in EMD simulations currently. We found that the phonon modes with mean free path larger than the system size are truncated in NEMD and contribute little to the overall thermal conductivity. The TDDDM provides direct physical origin for the well-known strong size effects in thermal conductivity prediction by NEMD.

cond-mat.mtrl-sci↗