Arrow Research search

Author name cluster

Xu Li

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

30 papers
2 author rows

Possible papers

30

AAAI Conference 2026 Conference Paper

A Logical Analysis of an Information Filtering Architecture Based on Epistemic Trust Inference

  • Xu Li
  • Leendert van der Torre
  • Liuwen Yu

In agent theory, epistemic trust is used to infer beliefs, for example by filtering out the information the agent receives from untrustworthy agents. Moreover, trust itself can be inferred from other information. We introduce a simple information filtering architecture that clearly distinguishes the relation between the two kinds of inference. We provide a logical analysis of the architecture, based on a new family of input/output logics. We then explore information filtering and belief manipulation within this formal framework. Our key finding is that with this architecture, some of the widely debated logical rules for trust inference are redundant with respect to information-filtering mechanisms and some others are redundant with respect to belief manipulation.

PDF Details DOI

EAAI Journal 2026 Journal Article

An enhanced you only look once model for multi-class apple detection in natural orchard environments

  • Xiaohang Liu
  • Zhao Zhang
  • Jiangfan Yu
  • Wanjia Hua
  • Xu Li
  • Han Li
  • Man Zhang
  • Chayan Kumer Saha

Multi-class apple detection can improve automatic apple-picking robots' efficiency. Existing studies classified apples into four occlusion types but struggled with clustered fruits and could not balance precision, speed, and model size. A robust Apple State You Only Look Once version 8 medium (AS-YOLOv8m) model was thus proposed for detecting apples into 11 classes according to the apples’ occlusion and clustering conditions. Core innovations included: (i) A cross-stage partial bottleneck module with the deformable convolution was designed to enhance feature extraction and geometric transformation modeling capabilities; (ii) the space-to-depth convolution module was embedded in the backbone network to improve small target detection; (iii) the large-target detection head was removed to lighten the model size; and (iv) the wise intersection over union box loss function was used to balance the loss of high- and low-quality anchor boxes. The model was trained (5, 845 images), validated (1, 948 images), and tested (1, 950 images) using 9, 743 apple images, which were augmented from 1, 149 original captures collected from commercial orchards under diverse lighting conditions. Results showed that AS-YOLOv8m achieved a higher mean average precision of 95. 8% in 11 classes than that of 95. 4% in 4 classes, which also outperformed other comparison models (<95. 1%) and prior research results (<91. 3%). The detection speed was 76. 9 frames per second, and the model size was 36. 2 megabytes. With its real-time capability, small model size, and high detection precision, the AS-YOLOv8m model stands as a promising multi-class apple detection method for the further improvement of robot picking effect and efficiency.

Details DOI

EAAI Journal 2026 Journal Article

Cascaded U-Net diffusion refiner for deformation prediction in hot strip rolling

  • Han Gao
  • Shanhong Cao
  • Xueqi Dong
  • Xu Li
  • Feng Luan
  • Dianhua Zhang

In the hot strip roughing process, Vertical-Horizontal rolling induces “fishtailing” deformation that leads to yield-reducing profile defects. While finite element method (FEM) simulations accurately model the complex elastoplastic deformation mechanisms underlying this phenomenon, their computational intensity impedes real-time process optimization and large-scale parametric analysis. To bridge this critical gap between accuracy and efficiency, we propose a Cascaded U-Net Diffusion Refiner (CUDR) framework for elastoplastic deformation prediction in hot rolling. The core design of this framework lies in the collaborative operation of two components: the U-Net first performs fast coarse prediction of deformation to provide a “warm start” foundation, and then the diffusion model conducts lightweight denoising refinement on the coarse prediction results. This refinement step primarily aims to suppress unphysical local fluctuations in the U-Net's predictions, thereby further enhancing the overall prediction precision. Validated on three orthogonal-sampled datasets with varying mesh resolutions, the CUDR reduces prediction errors by 26. 1% in Euclidean Mean Absolute Error and 28. 5% in Euclidean Mean Peak Absolute Error compared to the standalone U-Net. Moreover, Fourier-based spectral verification confirms that the framework suppresses unphysical local fluctuations. Critically, for fine-mesh cases, the CUDR achieves a 3900 times speedup over high-fidelity FEM simulations, making real-time deformation prediction feasible. This work demonstrates the substantial potential of generative diffusion models in advancing metal forming simulation, offering a new paradigm for balancing accuracy and efficiency in industrial manufacturing processes.

Details DOI

EAAI Journal 2026 Journal Article

Channel Clustering-based Attention Network for interpretable hard landing prediction

  • Hao Zhang
  • Huabo Sun
  • Yu Liu
  • Xinbin Zhao
  • Xu Li
  • Jiaxing Shang
  • Linjiang Zheng

Hard landing incidents are common flight safety events during the landing phase and are of significant concern in the aviation industry. Recent hard landing prediction methods tend to overemphasize temporal features while overlooking the landing process along the altitude dimension, importantly, altitude-based alignment enables more practical interpretability. Additionally, they often fail to capture dependencies between contributing factors and offer limited interpretability under fixed time windows. To address the above issues, we propose a Channel Clustering based Attention Network, termed CCAN, to predict hard landing incidents and identify their potential causes. Specifically, we resample and interpolate different flight parameters along the altitude dimension to align the landing process across different flights into a common reference frame for the subsequent interpretability. Subsequently, we design a channel clustering module that groups flight parameters into distinct clusters based on a predefined assignment threshold. Then, we employ graph attention network (GAT) to capture the dependencies between different flight parameters within and across clusters. To further reveal the interactions between flight parameters throughout the landing process, we incorporate attention mechanism into Gated Recurrent Units (GRUs) to extract informative temporal features. We conducted experiments on a real-world quick access recorders (QAR) dataset with 44, 729 Airbus A321 flights. Experimental results demonstrate that CCAN outperforms the baseline models in hard landing predictions and offers practical interpretability for hard landings by visualizing the dependencies between flight parameters and their interactions under the altitude reference during the landing process.

Details DOI

AAAI Conference 2026 Conference Paper

Dual-stream Relation-modeling Disentanglement for Cloth-Changing Person Re-Identification

  • Shijuan Huang
  • Hefei Ling
  • Zongyi Li
  • Xu Li
  • Zhao Lv

Cloth-changing person re-identification (CC-ReID) aims to identify individuals across non-overlapping cameras despite clothing variations. Existing methods are often constrained by two primary limitations: approaches using auxiliary modalities typically rely on a single specific cue, limiting their robustness, while feature disentanglement methods struggle with discrete labels that create inconsistencies between ground truth labels and modality semantic similarity. To overcome these limitations, we propose DRDnet, a unified framework that synergistically integrates dual auxiliary cues and advanced relation modeling. Specifically, our Dual-Stream Disentanglement (DSD) module leverages textual descriptions and parsing images to decouple clothing factors through high-level semantic supervision and pixel-level operations, yielding robust clothing-agnostic features. Simultaneously, our Modal Relation Modeling (MRM) module constructs feature memory banks and employs adaptive soft label smoothing, effectively enhancing image-text semantic alignment and reinforcing identity consistency across clothing changes. We evaluate DRDnet on several CC-ReID benchmarks to demonstrate its effectiveness and provide state-of-the-art performance across all benchmarks.

PDF Details DOI

AAAI Conference 2026 Conference Paper

Explore to Learn: Latent Exploration Through Disentangled Synergy Patterns for Reinforcement Learning in Overactuated Control

  • Yiming Wang
  • Kaiyan Zhao
  • Xu Li
  • Yan Li
  • Jiayu Chen
  • Steven Morad
  • Leong Hou U

Control in high-dimensional action spaces remains a fundamental challenge in reinforcement learning (RL), primarily due to inefficient exploration of the action space. While recent methods attempt to guide exploration, they often fall short of achieving the agility and coordination exhibited in biological motor control. Inspired by how organisms exploit muscle synergies for efficient movement, we propose Explore to Learn (ETL), a two-stage framework that first discovers fundamental synergy patterns and then leverages them for task-specific policy learning. In the first stage, ETL discovers underlying synergy patterns by deploying a targeted exploration policy. These patterns are modeled as latent directions in a low-dimensional space, along which the agent is guided to collect diverse and structured muscle activation trajectories. A variational autoencoder (VAE) is then trained to encode high-dimensional actions into a latent space whose dimensions correspond to the synergy patterns. In the second stage, the policy is trained entirely in this synergy-aware latent space, producing synergy coefficients that the decoder maps back to full-dimensional muscle actions. This structured representation significantly reduces the complexity of learning, while the decoder is further fine-tuned to enhance expressiveness and generalization across downstream tasks. Extensive experiments across musculoskeletal environments and the DMControl suite demonstrate that ETL consistently outperforms prior methods in both exploration efficiency and control performance, achieving superior scalability and generalization in overactuated control tasks.

PDF Details DOI

EAAI Journal 2026 Journal Article

Multivariate time series representation learning with multi-task graph neural network

  • Zhihui Gao
  • Baomin Xu
  • Jidong Yuan
  • Jinfeng Wang
  • Xu Li

Multivariate time series (MTS) representation learning poses a significant challenge in data mining. Current deep learning-based MTS representation methods mostly utilize neural networks to model temporal dependencies within individual univariate sequences, while failing to adequately consider the spatial relationships among different channels within MTS data. While a few methods leverage graph neural networks (GNNs) to model spatial dependencies, but they often do not effectively capture both global and local features simultaneously, potentially limiting the quality of MTS data representations. To overcome these limitations, we present MTGL, a novel Multi-Task Graph Neural Network-based MTS Representation Learning Framework. It leverages MTS reconstruction, global-level graph learning, and local-level graph learning to capture latent spatio-temporal dependencies without relying on predefined graph structures. To obtain global graph-level representations, MTGL performs message-passing and graph pooling operations, and simultaneously leverages a dynamic graph mechanism to capture associations across different windows for local-level representations. By fusing global and local features in a unified framework, MTGL effectively supports a variety of MTS tasks. Extensive experiments show that the proposed method outperforms existing state-of-the-art baselines on benchmark MTS datasets and the tunnel boring machine dataset.

Details DOI

EAAI Journal 2026 Journal Article

Opposition-based learning memetic algorithm for the maximum intersection of k -subsets problem

  • Wen Sun
  • Xu Li
  • Jin-Kao Hao
  • Wenlong Li
  • Zhipeng Lü

Given m elements and n subsets of elements, the maximum intersection of k -subsets (kMIS) problem is to select k subsets of elements to maximize the number of elements simultaneously covered by all of the selected subsets. As a general model, kMIS can be used to formulate some practical problems including data privacy control, community detection, and deoxyribonucleic acid microarray technology. This paper presents an opposition-based learning memetic algorithm that integrates opposition-based learning initialization, adaptive crossover, and solution-based tabu search. Experimental results on 608 instances show that the algorithm competes favorably with the state-of-the-art methods. The importance of the algorithmic components is experimentally validated.

Details DOI

EAAI Journal 2026 Journal Article

Sound zoning modulation within non-enclosed vehicular cavities: targeted compensation for dynamic reverberation and wind-induced noise

  • Xu Li
  • Xudong Wu
  • Jin Huang
  • Qitao Feng

Sound zoning facilitates the creation of multiple isolated listening regions within shared physical spaces, demonstrating considerable potential for personalized audio delivery. Nevertheless, practical implementation in vehicular cavities encounters distinctive challenges characterized by dynamically changing reverberant environments and accompanying wind-induced noise during window operation. To address this limitation, this study proposes a modulation methodology of sound zoning for non-enclosed vehicular cavities under varying window openings, with a focus on characterizing sound propagation with dynamic reverberation and wind-induced noise. A physics-informed neural network framework adapted to dynamic reverberant environments enables efficient and accurate prediction of acoustic transfer functions (ATFs) under varying window operations, circumventing the high cost of repeated simulations or measurements. Furthermore, by implementing tailored assignment of expected amplitude in acoustically shielded region, supported by established ATFs prediction, a targeted compensation strategy for sound zoning modulation adapted to non-enclosed vehicular cavities is developed. Eventually, to evaluate the effectiveness of sound zoning modulation in non-enclosed cavities, systematic validation is performed through reverberant sound propagation analysis, experimental ATFs measurements, and wind tunnel testing. Results demonstrate a root mean square error below 2 dB and prediction errors under 1 dB in low-frequency ranges, while maintaining consistent full-frequency performance. More significantly, the proposed compensation not only maintains acoustic privacy between regions but also reduces computational and measuring costs.

Details DOI

EAAI Journal 2026 Journal Article

Spatial Frequency Guidance Network for real-time lung computed tomography image segmentation

  • Mohammed A.M. Elhassan
  • JiaZhou Xiao
  • Bo Liu
  • Xu Li
  • Chenxi Huang
  • Jinbao Xie
  • Minglian Qiu

With the increasing demand for real-time lung tumor segmentation in clinical settings, there is a pressing need for efficient and accurate segmentation models. However, current state-of-the art methods are often burdened by high computational complexity and struggle to simultaneously capture fine-grained local details and global contextual cues essential for precise tumor delineation. To address these challenges, we present the Spatial Frequency Guidance Network (SFGNet), a lightweight encoder–decoder architecture tailored for real-time lung tumor segmentation. The proposed framework incorporates three key components: (1) a Residual Downsampling Block that reduces spatial resolution while preserving high-frequency structural cues; (2) a Spatial Frequency Guidance Module that integrates spatial and frequency-domain information to enhance multi-scale contextual representation; and (3) a Multi-Scale Feature Interaction Module that adaptively fuses features across resolutions to improve pixel-level classification accuracy. Experiments on two public lung computed tomography datasets demonstrate that the proposed architecture achieves a favorable balance between accuracy and efficiency. For example, the base variant attains best average across all metrics 97. 32% at real-time inference speed, while also producing best performance on a benchmark dataset of coronavirus disease (COVID-19) computed tomography scans. The source code, will be released publicly at: https: //github. com/mohamedac29/SFGNet.

Details DOI

EAAI Journal 2025 Journal Article

A Dual Two-Stage Attention-based Model for interpretable hard landing prediction from flight data

  • Jiaxing Shang
  • Xiaoquan Li
  • Ruixiang Zhang
  • Linjiang Zheng
  • Xu Li
  • Riquan Zhang
  • Xinbin Zhao
  • Fan Li

Hard landings are a significant safety concern in aviation, with potential consequences ranging from poor passenger experiences to serious injuries or fatalities. Predicting and explaining hard landing events are equally important for enhancing flight safety, the former makes it possible to give proactive warnings, while the latter helps pilots identify the reasons and refine their skills. However, existing studies generally lack a comprehensive consideration for the fine-grained characteristics of flight data containing both inter-temporal and inter-parametric relationships, resulting in suboptimal prediction performance. In addition, most of existing approaches aim at improving the prediction performance but fail to provide interpretability for the causes of hard landing. To address the above problems, we propose DUTSAM, a DUal Two-Stage Attention-based interpretable Model for hard landing prediction from quick access recorder (QAR) data. The model consists of dual parallel modules, each of which combines a convolutional feature encoder and a two-stage attention mechanism. The two encoders capture fine-grained characteristics by encoding multivariate data from temporal domain and parametric domain respectively. After that, the dual two-stage attention mechanism captures the inter-temporal and inter-parametric correlations in reverse order to predict hard landing and provide interpretation from both temporal and parametric perspectives. Experimental results on a real QAR dataset with 37, 920 flights show that DUTSAM achieves better prediction performance compared with other state-of-the-art baselines in terms of Precision, Recall, and F1-score. Additionally, case study demonstrates that DUTSAM can uncover key flight parameters and moments strongly correlated to the hard landing events.

Details DOI

FLAP Journal 2025 Journal Article

A Dynamic Logic of the Right to Know

  • Xu Li
  • Réka Markovich

Different meanings of the “right to know” can be distinguished based on the theory of normative positions. In this paper, we focus on one of them: the power to know. Intuitively, in a sender-receiver setting, the receiver’s power to know whether φ is the case means that the sender is obliged to (truthfully) announce the answer if the receiver asks the question φ? . Therefore, we de- velop a logic called LRK for reasoning about the power to know, the obligatory announcements, and the dynamics of questions and public announcements.

PDF Details

EAAI Journal 2025 Journal Article

A framework for compression index prediction considering geographical information and feature missing

  • Yuan-en Pang
  • Xu Li
  • Jin-peng Xin
  • Jia-ting Wang
  • Hao Cai

Using machine learning methods to predict the soil compression index (C c) has become a research hotspot due to its efficiency in computation and automatic data pattern mining. However, existing methods often overlook missing input features and fail to incorporate geographical information of the soil location, which limits data utilization efficiency and accuracy enhancement. To address these shortcomings, we propose a framework for C c prediction that considers geographical information and complements missing features. Firstly, geographical information, namely latitude (Lat), longitude (Lon), annual average maximum temperature (T max), and annual average minimum temperature (T min), is included according to the soil sample's collection site, as supplementary features in addition to liquid limit (LL), plasticity index (PI), natural water content (w), and initial void ratio (e 0) as adopted in the past studies. Subsequently, a missing feature completion method called “Mutual Prediction” is introduced to reduce the number of scenarios needing consideration for missing features. Finally, we construct a machine learning-based model named “Trunk-Branch” (TB model) to achieve C c prediction. The results show: (1) Incorporating geographical information increases the coefficient of determination (R 2) of C c from 0. 751 to 0. 901, with Lon contributing the most. (2) “Mutual Prediction” reduces the scenarios needing consideration for missing features from 15 to 3 and enhances the R 2 from 0. 746 to 0. 811. (3) Corresponding to these three missing scenarios, the TB model achieves R 2 values of 0. 612, 0. 720, and 0. 901, respectively.

Details DOI

IROS Conference 2025 Conference Paper

An Inflatable Deployable Origami Grasper for Adaptive and High-Load Grasping

  • Peng Yan
  • Guang Liang
  • Sen Wang
  • Hailin Huang
  • Wei Wang
  • Xu Li
  • Bing Li

Robotic graspers are essential for enhancing the efficiency and versatility of robots in grasping tasks. In this paper, we propose a novel inflatable deployable origami grasper with a rigid-flexible coupling structure. The proposed grasper can achieve multiple deployment configurations under a single pneumatic actuation, enabling both deployment and grasping operations while also allowing for passive self-folding during deflation. The design and fabrication of the grasper are presented. Then, the stiffness model for the inflatable deployable origami unit is developed based on the equivalent truss method. Experimental results show that the grasper successfully grasps objects of various shapes and sizes in both enveloping and fingertip grasping modes, using either two or four fingers. With its simple mechanical system and high deploy/fold ratio, the proposed grasper holds significant potential for applications in industrial automation and space exploration.

Details

YNICL Journal 2025 Journal Article

Association of iron deposition in MS lesion with remyelination capacity using susceptibility source separation MRI

  • Hyeong-Geol Shin
  • Woojun Kim
  • Jung Hwan Lee
  • Hyun-soo Lee
  • Yoonho Nam
  • Jiwoong Kim
  • Xu Li
  • Peter C.M. van Zijl

OBJECTIVES: signals within MS lesions using χ-separation and evaluate the association between lesional iron and remyelination capability. METHODS: signals. RESULTS: myelin signals (P < 0.001). After adjustment, lesions with early HPS demonstrated an annual loss in myelin signal (-1.94 ppb/year), whereas those without early HPS exhibited annual recovery (+0.66 ppb/year). Participants with confirmed disability improvement (CDI) had fewer HPS-positive lesions at baseline than those without CDI (P < 0.001). CONCLUSION: The presence of HPS is associated with impaired remyelination capacity and a lack of disease improvement in pwMS. Identifying HPS may help demarcate lesions more amenable to myelin repair therapies.

Details DOI

TMLR Journal 2025 Journal Article

Foundation Models Meet Federated Learning: A One-shot Feature-sharing Method with Privacy and Performance Guarantees

  • Mahdi Beitollahi
  • Alex Bie
  • Sobhan Hemati
  • Leo Maxime Brunswic
  • Xu Li
  • Xi Chen
  • Guojun Zhang

Adapting foundation models for downstream tasks via Federated Learning (FL) is a promising strategy for protecting privacy while leveraging the capability of foundation models. However, FL's iterative training and model transmission result in high communication costs and GPU memory demands, making large foundation models impractical for FL. This paper introduces a one-shot FL method with a server-side performance bound to enable foundation models by reducing communication costs and GPU memory requirements. Our approach, FedPFT (FL with Parametric Feature Transfer), involves clients learning and transferring parametric models for features extracted from frozen foundation models in a single round. Parametric models are then used to generate synthetic features at the server to train a classifier head. We evaluate FedPFT across eight vision datasets using three vision foundation models. Our findings demonstrate that FedPFT is agnostic to data heterogeneity and network topology and it enhances the communication-accuracy frontier up to 7.8\%. Finally, we show FedPFT's compatibility with differential privacy and its resilience against reconstruction attacks. Our work highlights the capability of private, feature-sharing methods for one-shot knowledge transfer using foundation models.

PDF Details

JAIR Journal 2025 Journal Article

From Knowledge to Action: Logics of Permitted and Obligatory Announcements

  • Xu Li
  • Guillaume Aucher
  • Dov Gabbay
  • Réka Markovich

We formalize the notions of “permitted and obligatory announcements” in the context of information security, such as privacy policy compliance. In a sender-receiver setting, we define the sender’s permitted and obligatory announcements in terms of the receiver’s ideal epistemic states (i.e., the epistemic states that comply with the given security policies). We propose two logics, LPOA and DLPOA, to reason about permitted and obligatory announcements in static and dynamic contexts, respectively. These two logics are completely axiomatized, and we also study generalizations in which the receiver’s knowledge is characterized by non-S5 logics. Our paper makes two main contributions to the formalization of permitted and obligatory announcements: First, we clarify the interplay between the sender’s permitted and obligatory announcements and the receiver’s knowledge. Second, we distinguish between weakly and strongly permitted announcements.

PDF Details DOI

IS Journal 2025 Journal Article

Wet-Neuromorphic Computing: A New Paradigm for Biological Artificial Intelligence

  • Jeewaka Perera
  • Sasitharan Balasubramaniam
  • Samitha Somathilaka
  • Qu Wen
  • Xu Li
  • Dharshana Kasthurirathna
  • Arman Roohi
  • M. Tyler Nelson

As we delve into a life governed by artificial intelligence (AI), ongoing research continues to discover new forms of intelligence that are efficient and closely mimic an organism’s brain in terms of performance. This article presents a new concept termed wet-neuromorphic computing, in which biological cells or organisms are leveraged to perform computational tasks using their natural molecular functions. We map key neuromorphic properties to natural biological computing observed in bacteria, 3-D organoids, and Caenorhabditis elegans. To expand beyond the inspiration of the brain to create conventional neuromorphic computing, the study presents a case study that demonstrates bacterial AI computing using the gene regulatory neural network derived from Escherichia coli’s gene regulatory network for pattern recognition, validated through wet lab experiments. Finally, challenges and future directions are discussed.

Details DOI

TMLR Journal 2024 Journal Article

DFML: Decentralized Federated Mutual Learning

  • Yasser H. Khalil
  • Amir Hossein Estiri
  • Mahdi Beitollahi
  • Nader Asadi
  • Sobhan Hemati
  • Xu Li
  • Guojun Zhang
  • Xi Chen

In the realm of real-world devices, centralized servers in Federated Learning (FL) present challenges including communication bottlenecks and susceptibility to a single point of failure. Additionally, contemporary devices inherently exhibit model and data heterogeneity. Existing work lacks a Decentralized FL (DFL) framework capable of accommodating such heterogeneity without imposing architectural restrictions or assuming the availability of additional data. To address these issues, we propose a Decentralized Federated Mutual Learning (DFML) framework that is serverless, supports nonrestrictive heterogeneous models, and avoids reliance on additional data. DFML effectively handles model and data heterogeneity through mutual learning, which distills knowledge between clients, and cyclically varying the amount of supervision and distillation signals. Extensive experimental results demonstrate consistent effectiveness of DFML in both convergence speed and global accuracy, outperforming prevalent baselines under various conditions. For example, with the CIFAR-100 dataset and 50 clients, DFML achieves a substantial increase of +17.20% and +19.95% in global accuracy under Independent and Identically Distributed (IID) and non-IID data shifts, respectively.

PDF Details

JBHI Journal 2023 Journal Article

Dynamics Combined With Hill Model for Functional Electrical Stimulation Ankle Angle Prediction

  • Xianghong Zhang
  • Ziqin Jiang
  • Xu Li
  • Pan Xu
  • Željka Lučev Vasić
  • Ivana Čuljak
  • Mario Cifrek
  • Min Du

Musculoskeletal models play an essential role in ankle rehabilitation research. The majority of the existing models have established the relationship between EMG and joint torque. However, EMG signal acquisition requires higher clinical conditions, such as sensitivity to external circumstances, motion artifacts and electrode position. To solve the nonlinear and time-varying nature of joint movement, a Functional Electrical Stimulation (FES) model was proposed in this study to simulate the whole process of ankle dorsiflexion. The model is combined with muscle contraction dynamics based on Hill model and ankle inverse dynamics to connect FES parameters, torques, and ankle angles. In addition, the extended Kalman filter (EKF) algorithm was applied to identify the unknown parameters of the model. Model validation experiment was performed by acquiring the actual data of healthy volunteers. Results showed that the root mean square error (RMSE) and normalized root mean square error (NRMSE) of this model were 11. 93%±0. 53% and 1. 39°±0. 26°, respectively, which means it can effectively predict the output variation of ankle joint angle while changing electrical stimulation parameters. Therefore, the proposed mode is essential for developing closed-loop feedback control of electrical stimulation and has the potential to help patients to conduct rehabilitation training.

Details DOI

YNIMG Journal 2023 Journal Article

Effect of motion, cortical orientation and spatial resolution on quantitative imaging of cortical R2* and magnetic susceptibility at 0.3 mm in-plane resolution at 7 T

  • Peter van Gelderen
  • Xu Li
  • Jacco A. de Zwart
  • Erin S. Beck
  • Serhat V. Okar
  • Yujia Huang
  • KuoWei Lai
  • Jeremias Sulam

MR images of the effective relaxation rate R2* and magnetic susceptibility χ derived from multi-echo T2*-weighted (T2*w) MRI can provide insight into iron and myelin distributions in the brain, with the potential of providing biomarkers for neurological disorders. Quantification of R2* and χ at submillimeter resolution in the cortex in vivo has been difficult because of challenges such as head motion, limited signal to noise ratio, long scan time, and motion related magnetic field fluctuations. This work aimed to improve the robustness for quantifying intracortical R2* and χ and analyze the effects from motion, spatial resolution, and cortical orientation. T2*w data was acquired with a spatial resolution of 0. 3 × 0. 3 × 0. 4 mm3 at 7 T and downsampled to various lower resolutions. A combined correction for motion and B0 changes was deployed using volumetric navigators. Such correction improved the T2*w image quality rated by experienced image readers and test-retest reliability of R2* and χ quantification with reduced median inter-scan differences up to 10 s−1 and 5 ppb, respectively. R2* and χ near the line of Gennari, a cortical layer high in iron and myelin, were as much as 10 s−1 and 10 ppb higher than the region at adjacent cortical depth. In addition, a significant effect due to the cortical orientation relative to the static field (B0) was observed in χ with a peak-to-peak amplitude of about 17 ppb. In retrospectively downsampled data, the capability to distinguish different cortical depth regions based on R2* or χ contrast remained up to isotropic 0. 5 mm resolution. This study highlights the unique characteristics of R2* and χ along the cortical depth at submillimeter resolution and the need for motion and B0 corrections for their robust quantification in vivo.

Details DOI

KR Conference 2022 Conference Paper

Dynamic Deontic Logic for Permitted Announcements

  • Xu Li
  • Dov Gabbay
  • Réka Markovich

In this paper, we introduce and study a dynamic deontic logic for permitted announcements. In our logic framework, it is permitted to announce something if announcing it would not lead to forbidden knowledge. It is shown that the logic is not compact, and we propose a sound and weakly complete Hilbert-style axiomatisation. We also study the computational complexity of the model checking problem and the decidability of the satisfiability problem. Finally, we introduce a neighbourhood semantics with a strongly complete axiomatisation.

PDF Details DOI

AIIM Journal 2022 Journal Article

Intelligent and strong robust CVS-LVAD control based on soft-actor-critic algorithm

  • Te Li
  • Wenbo Cui
  • Nan Xie
  • Heng Li
  • Haibo Liu
  • Xu Li
  • Yongqing Wang

Left ventricular assist device (LVAD) is an effective method to treat ventricular failure. According to the physiological conditions of different patients, the device adaptively adjusts its rotation speed to change LVAD output. In this study, a physiological control system for LVAD based on deep reinforcement learning (DRL) is proposed. The system estimates the amount of blood required by LVAD based on a Starling-like method. The DRL controller regulates LVAD to adjust the speed and quickly approach the target value. The changes of vascular resistance, myocardial contractility, and the transition from rest to exercise were simulated, and the single factor and mixed factor experiments were carried out to compare the effects of DRL controller and proportional integral derivative (PID) controller, which controls the system according to the difference between measured variables and expected values. Two metrics are used to illustrate the regulation effect: the sum of absolute error (SAE) and the response time of the two controllers, where SAE is the difference between the estimated required pumped blood flow LVADQ e and the actual measured blood flow LVADQ m. The experimental result shows that the SAE of the DRL controller is 47. 6% of that of the PID controller, and the response time of the DRL controller is 38. 6% of that of the PID controller. This study demonstrates that the LVAD based on the DRL controller can respond more quickly and more effectively to the different physiological needs of a variety of patients than a PID controller.

Details DOI

JBHI Journal 2022 Journal Article

Regularized Asymmetric Susceptibility Tensor Imaging in the Human Brain in Vivo

  • Yuting Shi
  • Steven Cao
  • Xu Li
  • Ruimin Feng
  • Jie Zhuang
  • Yuyao Zhang
  • Chunlei Liu
  • Hongjiang Wei

Susceptibility tensor imaging (STI) is a promising tool for studying orientation-dependent tissue magnetic susceptibility and for mapping white matter fiber orientations complementary to diffusion tensor imaging (DTI). However, the limited head rotation range within modern head coils for data acquisition makes in vivo STI reconstruction ill-conditioned. Conventional STI reconstruction method is usually vulnerable to noise and requires sufficiently large head rotations to solve this ill-conditioned inverse problem. In this study, based on the recently proposed asymmetric STI (aSTI) model, a new method termed aSTI+ was proposed to improve in vivo STI reconstruction by enforcing isotropic susceptibility tensor inside cerebrospinal fluid (CSF) and applying morphology constraint in white matter. Experimental results showed superior performance of the proposed method with reduced noise, improved tissue contrast and better fiber orientation estimation over previous methods. Thus aSTI+ may promote in vivo human brain STI studies on white matter and myelin-related brain diseases.

Details DOI

AAAI Conference 2020 Conference Paper

Meta-CoTGAN: A Meta Cooperative Training Paradigm for Improving Adversarial Text Generation

  • Haiyan Yin
  • Dingcheng Li
  • Xu Li
  • Ping Li

Training generative models that can generate high-quality text with sufficient diversity is an important open problem for Natural Language Generation (NLG) community. Recently, generative adversarial models have been applied extensively on text generation tasks, where the adversarially trained generators alleviate the exposure bias experienced by conventional maximum likelihood approaches and result in promising generation quality. However, due to the notorious defect of mode collapse for adversarial training, the adversarially trained generators face a quality-diversity trade-off, i. e. , the generator models tend to sacrifice generation diversity severely for increasing generation quality. In this paper, we propose a novel approach which aims to improve the performance of adversarial text generation via efficiently decelerating mode collapse of the adversarial training. To this end, we introduce a cooperative training paradigm, where a language model is cooperatively trained with the generator and we utilize the language model to efficiently shape the data distribution of the generator against mode collapse. Moreover, instead of engaging the cooperative update for the generator in a principled way, we formulate a meta learning mechanism, where the cooperative update to the generator serves as a high level meta task, with an intuition of ensuring the parameters of the generator after the adversarial update would stay resistant against mode collapse. In the experiment, we demonstrate our proposed approach can efficiently slow down the pace of mode collapse for the adversarial text generators. Overall, our proposed method is able to outperform the baseline approaches with significant margins in terms of both generation quality and diversity in the testified domains.

PDF Details

AAAI Conference 2019 Conference Paper

Multi-Agent Discussion Mechanism for Natural Language Generation

  • Xu Li
  • Mingming Sun
  • Ping Li

We introduce the discussion mechanism into the multiagent communicating encoder-decoder architecture for Natural Language Generation (NLG) tasks and prove that by applying the discussion mechanism, the communication between agents becomes more effective. Generally speaking, an encoder-decoder architecture predicts target-sequence word by word in several time steps. At each time step of prediction, agents with the discussion mechanism predict the target word after several discussion steps. In the first step of discussion, agents make their choice independently and express their decision to other agents. In the next discussion step, agents collect other agents’ decision to update their own decisions, then express the updated decisions to others again. After several iterations, the agents make their final decision based on a well-communicated situation. The benefit of the discussion mechanism is that multiple encoders can be designed as different structures to fit the specified input or to fetch different representations of inputs. We train and evaluate the discussion mechanism on Table to Text Generation, Text Summarization and Image Caption tasks, respectively. Our empirical results demonstrate that the proposed multi-agent discussion mechanism is helpful for maximizing the utility of the communication between agents.

PDF Details

YNIMG Journal 2019 Journal Article

Multi-atlas tool for automated segmentation of brain gray matter nuclei and quantification of their magnetic susceptibility

  • Xu Li
  • Lin Chen
  • Kwame Kutten
  • Can Ceritoglu
  • Yue Li
  • Ningdong Kang
  • John T. Hsu
  • Ye Qiao

Quantification of tissue magnetic susceptibility using MRI offers a non-invasive measure of important tissue components in the brain, such as iron and myelin, potentially providing valuable information about normal and pathological conditions during aging. Despite many advances made in recent years on imaging techniques of quantitative susceptibility mapping (QSM), accurate and robust automated segmentation tools for QSM images that can help generate universal and sharable susceptibility measures in a biologically meaningful set of structures are still not widely available. In the present study, we developed an automated process to segment brain nuclei and quantify tissue susceptibility in these regions based on a susceptibility multi-atlas library, consisting of 10 atlases with T1-weighted images, gradient echo (GRE) magnitude images and QSM images of brains with different anatomic patterns. For each atlas in this library, 10 regions of interest in iron-rich deep gray matter structures that are better defined by QSM contrast were manually labeled, including caudate, putamen, globus pallidus internal/external, thalamus, pulvinar, subthalamic nucleus, substantia nigra, red nucleus and dentate nucleus in both left and right hemispheres. We then tested different pipelines using different combinations of contrast channels to bring the set of labels from the multi-atlases to each target brain and compared them with the gold standard manual delineation. The results showed that the segmentation accuracy using dual contrasts QSM/T1 pipeline outperformed other dual-contrast or single-contrast pipelines. The dice values of 0. 77 ± 0. 09 using the QSM/T1 multi-atlas pipeline rivaled with the segmentation reliability obtained from multiple evaluators with dice values of 0. 79 ± 0. 07 and gave comparable or superior performance in segmenting subcortical nuclei in comparison with standard FSL FIRST or recent multi-atlas package of volBrain. The segmentation performance of the QSM/T1 multi-atlas was further tested on QSM images acquired using different acquisition protocols and platforms and showed good reliability and reproducibility with average dice of 0. 79 ± 0. 08 to manual labels and 0. 89 ± 0. 04 in an inter-protocol manner. The extracted quantitative magnetic susceptibility values in the deep gray matter nuclei also correlated well between different protocols with inter-protocol correlation constants all larger than 0. 97. Such reliability and performance was ultimately validated in an external dataset acquired at another study site with consistent susceptibility measures obtained using the QSM/T1 multi-atlas approach in comparison to those using manual delineation. In summary, we designed a susceptibility multi-atlas tool for automated and reliable segmentation of QSM images and for quantification of magnetic susceptibilities. It is publicly available through our cloud-based platform (www. mricloud. org). Further improvement on the performance of this multi-atlas tool is expected by increasing the number of atlases in the future.

Details DOI

YNIMG Journal 2014 Journal Article

Quantitative magnetic susceptibility mapping without phase unwrapping using WASSR

  • Issel Anne L. Lim
  • Xu Li
  • Craig K. Jones
  • Jonathan A.D. Farrell
  • Deepti S. Vikram
  • Peter C.M. van Zijl

The magnetic susceptibility of tissue within and around an image voxel affects the magnetic field and thus the local frequency in that voxel. Recently, it has been shown that spatial maps of frequency can be used to quantify local susceptibility if the contributions of surrounding tissue can be deconvolved. Currently, such quantitative susceptibility mapping (QSM) methods employ gradient recalled echo (GRE) imaging to measure spatial differences in the signal phase evolution as a function of echo time, from which frequencies can be deduced. Analysis of these phase images, however, is complicated by phase wraps, despite the availability and usage of various phase unwrapping algorithms. In addition, lengthy high-resolution GRE scanning often heats the magnet bore, causing the magnetic field to drift over several Hertz, which is on the order of the frequency differences between tissues. Here, we explore the feasibility of applying the WAter Saturation Shift Referencing (WASSR) method for 3D whole brain susceptibility imaging. WASSR uses direct saturation of water protons as a function of frequency irradiation offset to generate frequency maps without phase wraps, which can be combined with any image or spectroscopy acquisition. By utilizing a series of fast short-echo-time direct saturation images with multiple radiofrequency offsets, a frequency correction for field drift can be applied based on the individual image phases. Regions of interest were delineated with an automated atlas-based method, and the average magnetic susceptibilities calculated from frequency maps obtained from WASSR correlated well with those from the phase-based multi-echo GRE approach at 3T.

Details DOI

YNIMG Journal 2013 Journal Article

Human brain atlas for automated region of interest selection in quantitative susceptibility mapping: Application to determine iron content in deep gray matter structures

  • Issel Anne L. Lim
  • Andreia V. Faria
  • Xu Li
  • Johnny T.C. Hsu
  • Raag D. Airan
  • Susumu Mori
  • Peter C.M. van Zijl

The purpose of this paper is to extend the single-subject Eve atlas from Johns Hopkins University, which currently contains diffusion tensor and T1-weighted anatomical maps, by including contrast based on quantitative susceptibility mapping. The new atlas combines a “deep gray matter parcellation map” (DGMPM) derived from a single-subject quantitative susceptibility map with the previously established “white matter parcellation map” (WMPM) from the same subject's T1-weighted and diffusion tensor imaging data into an MNI coordinate map named the “Everything Parcellation Map in Eve Space, ” also known as the “EvePM. ” It allows automated segmentation of gray matter and white matter structures. Quantitative susceptibility maps from five healthy male volunteers (30 to 33years of age) were coregistered to the Eve Atlas with AIR and Large Deformation Diffeomorphic Metric Mapping (LDDMM), and the transformation matrices were applied to the EvePM to produce automated parcellation in subject space. Parcellation accuracy was measured with a kappa analysis for the left and right structures of six deep gray matter regions. For multi-orientation QSM images, the Kappa statistic was 0. 85 between automated and manual segmentation, with the inter-rater reproducibility Kappa being 0. 89 for the human raters, suggesting “almost perfect” agreement between all segmentation methods. Segmentation seemed slightly more difficult for human raters on single-orientation QSM images, with the Kappa statistic being 0. 88 between automated and manual segmentation, and 0. 85 and 0. 86 between human raters. Overall, this atlas provides a time-efficient tool for automated coregistration and segmentation of quantitative susceptibility data to analyze many regions of interest. These data were used to establish a baseline for normal magnetic susceptibility measurements for over 60 brain structures of 30- to 33-year-old males. Correlating the average susceptibility with age-based iron concentrations in gray matter structures measured by Hallgren and Sourander (1958) allowed interpolation of the average iron concentration of several deep gray matter regions delineated in the EvePM.

Details DOI

YNIMG Journal 2012 Journal Article

Mapping magnetic susceptibility anisotropies of white matter in vivo in the human brain at 7T

  • Xu Li
  • Deepti S. Vikram
  • Issel Anne L. Lim
  • Craig K. Jones
  • Jonathan A.D. Farrell
  • Peter C.M. van Zijl

High-resolution magnetic resonance phase- or frequency-shift images acquired at high field show contrast related to magnetic susceptibility differences between tissues. Such contrast varies with the orientation of the organ in the field, but the development of quantitative susceptibility mapping (QSM) has made it possible to reproducibly image the intrinsic tissue susceptibility contrast. However, recent studies indicate that magnetic susceptibility is anisotropic in brain white matter and, as such, needs to be described by a symmetric second-rank tensor ( χ ¯ ¯ ). To fully determine the elements of this tensor, it would be necessary to acquire frequency data at six or more orientations. Assuming cylindrical symmetry of the susceptibility tensor in myelinated white matter fibers, we propose a simplified method to reconstruct the susceptibility tensor in terms of a mean magnetic susceptibility, M M S = ( χ // + 2 χ ⊥ ) / 3 and a magnetic susceptibility anisotropy, M S A = χ / / − χ ⊥, where χ // and χ ⊥ are susceptibility parallel and perpendicular to the white matter fiber direction, respectively. Computer simulations show that with a practical head rotation angle of around 20°–30°, four head orientations suffice to reproducibly reconstruct the tensor with good accuracy. We tested this approach on whole brain 1×1×1mm3 frequency data acquired from five healthy subjects at 7T. The frequency information from phase images collected at four head orientations was combined with the fiber direction information extracted from diffusion tensor imaging (DTI) to map the white matter susceptibility tensor. The MMS and MSA were quantified for regions in several large white matter fiber structures, including the corona radiata, posterior thalamic radiation and corpus callosum. MMS ranged from −0. 037 to −0. 053ppm (referenced to CSF being about zero). MSA values could be quantified without the need for a reference and ranged between 0. 004 and 0. 029ppm, in line with the expectation that the susceptibility perpendicular to the fiber is more diamagnetic than the one parallel to it.

Details DOI
v2026.09.13