Author name cluster

Zhi Yang

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.

31 papers

2 author rows

JBHI Journal 2026 Journal Article

Adaptive Spectral Graph Attention Filtering Network for Alzheimer's Disease Classification Using Multimodal Data

Zhi Yang
Bo Cheng
Haitao Gan
Zhongwei Huang
Ran Zhou
Ji Wang

Early detection of Alzheimer's Disease (AD) is critical for timely intervention and management. However, existing graph-based approaches often fail to fully leverage the rich spectral-domain information inherent in brain network signals. To address this limitation, we propose an Adaptive Spectral Graph Attention Filtering Network (ASGAFN), which effectively models the spectral structures of functional and structural brain networks to enhance classification performance. Specifically, we first construct structural and functional brain network graphs from diffusion tensor imaging (DTI) and resting-state functional magnetic resonance imaging (rs-fMRI). Subsequently, a frequency-encoding-guided attention mechanism is designed to learn a shared spectral response function across graphs. This enables the construction of interpretable and adaptive spectral filters while mitigating semantic misalignment across different spectral domains. Furthermore, a spectral energy sensing module is incorporated to facilitate graph-specific adaptation, thereby enhancing flexibility and subject-level discriminability. Finally, the refined spectral signals are transformed back to the spatial domain and fused via a Multimodal Fusion and Enhancement Layer (MFEL). Extensive experiments demonstrate that ASGAFN significantly outperforms multiple baselines in AD-related classification tasks. It achieves accuracies of 96. 64% (AD vs. NC), 90. 48% (MCI vs. NC), and 91. 75% (AD vs. MCI). Additionally, it attains an accuracy of 87. 12% in the three-class classification task, underscoring its effectiveness in distinguishing among multiple disease stages. These findings validate the efficacy of spectral-domain modeling and multimodal fusion.

YNIMG Journal 2026 Journal Article

Frequency-dependent modulation of human reward circuitry: A comparative study of theta, gamma, and high-frequency temporal interference

Yongxi Zhang
Zhenxiang Zang
Rui Liu
Ke Liu
Gang Wang
Zhi Yang

BACKGROUND: Temporal interference (TI) stimulation offers a noninvasive neuromodulation technique for targeting deep brain structures while sparing overlying cortical tissue. While early applications have validated TI's capacity to engage subcortical targets such as the hippocampus and striatum, the frequency-dependent mechanisms governing its efficacy remain poorly understood. This is particularly critical for the nucleus accumbens (NAc), a key hub in reward circuitry where invasive deep brain stimulation (DBS) typically operates at high frequencies (∼130 Hz). METHODS: In this study, we investigated whether TI stimulation induces frequency-specific modulation of NAc activity and its functional coupling with the prefrontal cortex. Using a within-subject, counterbalanced design, we applied individualized NAc-targeted TI stimulation at three distinct envelope frequencies (5 Hz, 40 Hz, and 130 Hz) in 24 healthy adults. Resting-state fMRI was acquired pre- and post-stimulation. RESULTS: Results revealed a distinct dissociation between local and circuit-level effects: TI stimulation induced no statistically significant changes in local spontaneous activity within the NAc across any frequency condition. In contrast, 130 Hz stimulation selectively reduced functional connectivity between the NAc and the medial prefrontal cortex (mPFC), whereas 5 Hz and 40 Hz conditions produced no such effect. Notably, despite the absence of significant group-level local modulation, the magnitude of individual NAc activity reduction under 130 Hz stimulation was significantly correlated with the extent of NAc-mPFC decoupling (r = -0.53). Exploratory analyses further revealed increased activity in the adjacent dorsal striatum (right putamen), consistent with a conduction-block model at the target core. CONCLUSIONS: These findings suggest that high-frequency TI mimics the network-disrupting effects of high-frequency DBS, offering evidence that TI can noninvasively modulate deep reward circuits in a parameter-specific manner for potential clinical application.

EAAI Journal 2026 Journal Article

Multi-angle feature enhancement for multi-defect category insulator defect detection in the wild

Zhi Yang
Zhiqing Guo
Liejun Wang

Small insulator defects in complex backgrounds are poorly detected during inspections by unmanned aerial vehicles (UAVs) on transmission lines. Meanwhile, many insulator datasets have a single type of insulator defect, making it difficult to meet the requirements of practical inspection tasks. To address the above issues, we propose a more balanced insulator defect detection network (BDD-Net). First, we propose a multi-angle feature enhancement module (MAFE), which allows for a larger receptive field, enabling the network to accurately localize and identify small-sized insulator defects. Next, we construct a new multi-scale bidirectional feature fusion module (MBFF) for more efficient and faster feature fusion. In addition, we introduce wise intersection over union (WIoU) as the localization loss function of BDD-Net to reduce the impact of low-quality samples on training. Finally, we craft a dataset containing one insulator string category and four different insulator defect categories. The experimental results demonstrate that BDD-Net achieves a 1 percentage point improvement in detection accuracy over the baseline, reaching 98. 2%. Meanwhile, with significantly reduced parameters and computational costs, BDD-Net attains 52. 3% faster detection speed. This optimized balance between detection accuracy and processing efficiency makes BDD-Net particularly suitable for practical transmission line insulator defect inspection. The dataset can be found on GitHub: https: //github. com/1wsxiao/BDD-Net.

EAAI Journal 2025 Journal Article

A multi-Layered neural control framework: Combining central pattern generators utilizing muscle synergy Theory and incorporating proprioceptors

Fuhao Mo
Zhi Yang
Yiran Cui
Dongxi Ma
Haining Wang

Understanding human control strategy can promote the application of neuroscience in rehabilitation medicine, and drive the development of bionic robotics. The current research is to develop a bionic human control framework including both upper-level control signal from the central nervous system and lower feedback control loop. The upper-level control strategy is a combination of Central pattern generators (CPGs) signals based on the principle of muscle synergy, the weight of which is adaptively adjusted by reinforcement learning. The lower-level control involves feedback signals from primary proprioceptors. A neuromuscular arm model was established by integrating the bionic control strategy in an upper limb musculoskeletal dynamics model. The model feasibility was evaluated by elbow flexion experiments in the human sagittal plane and trajectory planning experiments in three-dimensional (3D) space. The indices such as joint rotation angles, muscle activation levels, and co-correlation index (CCI) were adopted. The simulated joint angles are well correlated with the experimental results of the two types, while the acceptable results for muscle activation levels and CCI values can only be obtained in the elbow flexion experiment. The proprioceptive feedback loop is also indicated to be necessary for accurate and stable human movement prediction. In summary, the current control strategy can well accommodate complex joint kinematics, and be further improved by addressing coordination issues of multiple muscle activations.

EAAI Journal 2025 Journal Article

Adaptive feature selection with flexible mapping for diagnosis and prediction of Parkinson's disease

Zhongwei Huang
Jianqiang Li
Jiatao Yang
Jun Wan
Jianxia Chen
Zhi Yang
Ming Shi
Ran Zhou

Parkinson's disease (PD) is a neurodegenerative disorder in which symptoms gradually worsen over time, with no cure currently available. Therefore, its early-stage treatment is crucial. However, medical neuroimaging data often contain redundant features and high dimensionality, which can negatively impact algorithm accuracy and require greater computational resources. To address this challenge, a supervised feature selection algorithm is proposed for the early diagnosis of PD. Specifically, the proposed method incorporates adaptive learning during iterations, which enables adaptive updating of the similarity matrix and selection of informative features. Meanwhile, flexible mapping is introduced to address the limitations of strict linear mapping. To address the optimization challenge of the model, we propose an alternative iterative algorithm and provide theoretical proof of its strict convergence. To measure the effectiveness of the algorithm, we conducted experiments on the Parkinson's Progression Markers Initiative (PPMI) public dataset, involving three groups: PD vs. normal controls (NC), scans without evidence of dopaminergic deficit (SWEDD) vs. NC, and PD vs. SWEDD. Based on baseline data, the accuracies for these three groups were 83. 29%, 87. 67%, and 85. 16%, respectively. Using the 12-month data, the accuracies improved to 79. 70%, 96. 79%, and 96. 95%, while the 24-month data resulted in accuracies of 91. 33%, 70. 10%, and 93. 36%. The experimental results demonstrate that the proposed method outperforms existing feature selection methods in overall performance.

EAAI Journal 2025 Journal Article

ESIGCF: Extremely simplified but intent-enhanced graph collaborative filtering for recommendation

Zhi Yang
Ruizhang Huang
Yanping Chen
Chuan Lin
Yongbin Qin

Graph convolutional networks (GCNs) and graph contrastive learning (GCL) have substantially advanced recommender systems by modeling high-order user–item interactions and leveraging self-supervised signals. However, many existing methods overemphasize user–user or item–item similarities and rely on complex intent modeling, leading to increased complexity and limited exposure to diverse items. To address these challenges, we propose ESIGCF (Extremely Simplified but Intent-enhanced Graph Collaborative Filtering) — a lightweight yet effective recommendation framework. ESIGCF explicitly defines user intent as the inner product between user and item embedding vectors and comprises two primary modules: (i) an intent-enhanced GCN that uses hybrid normalization (combining mean- and symmetric-normalization) to capture fine-grained user–item preferences without additional intent parameters, and (ii) an intent-aware GCL that aligns user–item pairs and positive and generated negative items. Negative samples are generated via a non-linear activation of item embedding interactions, promoting exposure to varied candidates without data augmentation. Experiments on three public datasets (Alibaba-iFashion, Yelp2018, Amazon-Book) show that ESIGCF consistently outperforms state-of-the-art baselines. For instance, on Alibaba-iFashion, ESIGCF achieves Recall@20 of 0. 1273 versus 0. 1059 for the best intent-enhanced baseline (a 20. 2% relative improvement). Comprehensive experiments confirm that ESIGCF effectively captures latent user intent, mitigates popularity bias, and enhances recommendation performance with reduced complexity. Our code is available at https: //github. com/Yangzhi22/ESIGCF.

JBHI Journal 2025 Journal Article

Incomplete Multi-view Data Learning via Adaptive Embedding and Partial l 2,1 Norm Constraints for Parkinson's Disease Diagnosis

Zhongwei Huang
Kai Wang
Chao Chen
Jianxia Chen
Jun Wan
Zhi Yang
Ran Zhou
Haitao Gan

Parkinson's disease (PD) is a progressive neurodegenerative disorder characterized by mental abnormalities and motor dysfunction. Its early classification and prediction of clinical scores have been major concerns for researchers. Currently, multi-view data learning has become an essential research area due to the capacity of multiple views to provide complementary insights from various perspectives. However, the discontinuous distribution, data missing complexity, small sample size, and redundant features in multi-view datasets pose a substantial obstacle, and most existing multi-view learning methods are unable to handle these challenges effectively. In this study, we propose a novel incomplete multi-view data learning framework (IMVDL) via dynamic embedding and partial l 2, 1 norm constraints for PD diagnosis. Specifically, multi-view dynamic embedding can adapt to any view missing scene, thereby linearly/nonlinearly mapping incomplete multi-view data to low-dimensional manifold spaces and generating complete multi-view data representations. The partial l 2, 1 norm constraint can ignore larger feature weight values and perform l 2, 1 norm sparse on the remaining weights, thereby avoiding the sparse bias problem caused by larger weight values. An efficient iterative algorithm is derived to find the optimal solution of the IMVDL method. We conduct extensive experiments using multi-modal neuroimage data from the Parkinson's Progression Markers Initiative (PPMI) database. The results demonstrate that the IMVDL method is superior to other comparative methods. The source code for IMVDL is available at https://github.com/a610lab/IMVDL/.

YNICL Journal 2025 Journal Article

Language dysfunction and related amyloid-β in Alzheimer’s disease mediated by brain functional connectivity changes

Jin-Wen Xiao
Jin Wang
Jin-Tao Wang
Hai-Xia Li
Jian-Ping Li
Xin-Yi Xie
Jie-Li Geng
Nan Zhi

BACKGROUND: Language dysfunction occurs early in Alzheimer's disease (AD) and whether amyloid-β pathology is related to language dysfunction remains unclear. Functional connectivity (FC) in language networks is critical for language function. We hypothesize that altered FCs in fronto-temporal regions (core language areas) mediate the association between amyloid-β burden and language impairment. METHODS: A total of 110 individuals were recruited including 44 cognitively unimpaired individuals, 24 mild cognitive impairment patients and 42 AD patients. The acoustic features and semantic content were extracted from speech recordings of cookie-theft picture description for language assessment. Resting-state functional magnetic resonance imaging and 18F-florbetapir positron emission tomography were conducted to estimate fronto-temporal FCs and amyloid-β burden. RESULTS: The acoustic features of average duration of silence segments, percentage of silence duration and ratio of hesitation/speech counts were significantly longer in AD patients, which were correlated with the semantic content. Among them, the average duration of silence segments was positively correlated with global amyloid-β burden (r = 0.30, p = 0.015). The fronto-temporal FCs decreased in AD patients. Mediation analysis revealed that the reduced FC between the left rostroposterior superior temporal sulcus (rpSTS) and the right dorsal inferior frontal gyrus mediated the association between global amyloid-β burden and average duration of silence segments (a*b = 0.19, p = 0.026). Additionally, it also mediated the effect of amyloid-β burden in the left rpSTS region on average duration of silence segments (a*b = 0.15, p = 0.030). CONCLUSIONS: Our findings showed that reduced fronto-temporal FCs mediated the association between amyloid-β burden and language dysfunction and further demonstrated the vulnerability of the left rpSTS in response to amyloid-β burden, which might lead to decreased FC with other regions.

YNICL Journal 2024 Journal Article

Fine hippocampal morphology analysis with a multi-dataset cross-sectional study on 2911 subjects

Qinzhu Yang
Guojing Chen
Zhi Yang
Tammy Riklin Raviv
Yi Gao

CA1 subfield and subiculum of the hippocampus contain a series of dentate bulges, which are also called hippocampus dentation (HD). There have been several studies demonstrating an association between HD and brain disorders. Such as the number of hippocampal dentation correlates with temporal lobe epilepsy. And epileptic hippocampus have a lower number of dentation compared to contralateral hippocampus. However, most studies rely on subjective assessment by manual searching and counting in HD areas, which is time-consuming and labor-intensive to process large amounts of samples. And to date, only one objective method for quantifying HD has been proposed. Therefore, to fill this gap, we developed an automated and objective method to quantify HD and explore its relationship with neurodegenerative diseases. In this work, we performed a fine-scale morphological characterization of HD in 2911 subjects from four different cohorts of ADNI, PPMI, HCP, and IXI to quantify and explore differences between them in MR T1w images. The results showed that the degree of right hippocampal dentation are lower in patients with Alzheimer's disease than samples in mild cognitive impairment or cognitively normal, whereas this change is not significant in Parkinson's disease progression. The innovation of this paper that we propose a quantitative, robust, and fully automated method. These methodological innovation and corresponding results delineated above constitute the significance and novelty of our study. What's more, the proposed method breaks through the limitations of manual labeling and is the first to quantitatively measure and compare HD in four different brain populations including thousands of subjects. These findings revealed new morphological patterns in the hippocampal dentation, which can help with subsequent fine-scale hippocampal morphology research.

EAAI Journal 2024 Journal Article

WAL-Net: Weakly supervised auxiliary task learning network for carotid plaques classification

Haitao Gan
Lingchao Fu
Ran Zhou
Weiyan Gan
Furong Wang
Xiaoyan Wu
Zhi Yang
Zhongwei Huang

The classification of carotid artery ultrasound images is a crucial step for diagnosing carotid plaques, holding significant clinical relevance for predicting the risk of stroke. Recent research suggests that utilizing supervised plaque segmentation as an auxiliary task for classification can enhance performance by leveraging the correlation between segmentation and classification tasks. However, this approach relies on obtaining a substantial amount of challenging-to-acquire segmentation annotations. This paper proposes a novel weakly supervised auxiliary task learning network model (WAL-Net) to explore the interdependence between carotid plaque classification and segmentation tasks. The plaque classification task is the primary one, while the plaque segmentation task serves as the auxiliary one, providing valuable information to enhance the performance of the primary task. Weakly supervised learning is adopted in the auxiliary task in which the segmentation masks are not provided. Experiments and evaluations are conducted on a dataset comprising 1270 carotid plaque ultrasound images from Wuhan University Zhongnan Hospital. Results indicate that WAL-Net achieved an approximately 1. 3% improvement in carotid plaque classification accuracy compared to the baseline network. Specifically, the accuracy of mixed-echoic plaques classification increased by approximately 3. 3%, demonstrating the effectiveness of WAL-Net.

IJCAI Conference 2024 Conference Paper

X-former Elucidator: Reviving Efficient Attention for Long Context Language Modeling

Xupeng Miao
Shenhan Zhu
Fangcheng Fu
Ziyu Guo
Zhi Yang
Yaofeng Tu
Zhihao Jia
Bin Cui

Transformer-based LLMs are becoming increasingly important in various AI applications. However, apart from the success of LLMs, the explosive demand of long context handling capabilities is a key and in-time problem for both academia and industry. Due to the limitations from the quadratic complexity of the attention mechanism, long context scenarios require much more resources for LLM development and deployment, bringing huge challenges to the underlying AI infrastructure. Meanwhile, we observe that there is a trend of reviving previous efficient attention mechanisms to latest LLMs. However, it still remains an open question about how to select from these diverse approaches in practice. In this paper, we answer this question from several aspects. First, we revisit these latest long-context LLM innovations and discuss their relationship with prior approaches with a novel and comprehensive taxonomy. Next, we conduct a thorough evaluation over various types of workloads considering both efficiency and effectiveness. Finally, we provide an in-depth analysis, summarize our key findings, and offer insightful suggestions on the trade-offs of designing and deploying efficient attention mechanisms for Transformer-based LLMs.

PDF Details DOI

EAAI Journal 2023 Journal Article

Safe semi-supervised clustering based on Dempster–Shafer evidence theory

Haitao Gan
Zhi Yang
Ran Zhou
Li Guo
Zhiwei Ye
Rui Huang

In this paper, we propose a safe semi-supervised clustering algorithm based on Dempster–Shafer (D–S) evidence theory. The motivation is that D–S evidence theory can be used to fuse multiple base clustering results and obtain robust confidence estimations of mislabeled samples. Firstly, the proposed algorithm constructs multiple base clusters using fuzzy c -means and kernel fuzzy c -means. Base clusters with good performance are selected according to a clustering validity function. Then, D–S evidence theory is used to fuse the results of the selected base clusters, and the confidence of labeled samples is calculated based on the fused results. Finally, we construct a p -nearest neighbor graph to limit the outputs of labeled samples with low confidence to be those of the p nearest unlabeled samples. It is desired to reduce the negative influence of labeled samples with low confidence and achieve safe exploitation. To verify the effectiveness of the proposed algorithm, we compare it to several unsupervised and semi-supervised clustering algorithms. The results demonstrate that our algorithm yields higher accuracy and is more stable.

YNIMG Journal 2022 Journal Article

Growth charts of brain morphometry for preschool children

Hongxi Zhang
Jia Li
Xiaoli Su
Yang Hu
Tianmei Liu
Shaoqing Ni
Haifeng Li
Xi-Nian Zuo

Brain development from 1 to 6 years of age anchors a wide range of functional capabilities and carries early signs of neurodevelopmental disorders. However, quantitative models for depicting brain morphology changes and making individualized inferences are lacking, preventing the identification of early brain atypicality during this period. With a sample size of 285, we characterized the age dependence of the cortical thickness and subcortical volume in neurologically normal children and constructed quantitative growth charts of all brain regions for preschool children. While the cortical thickness of most brain regions decreased with age, the entorhinal and parahippocampal regions displayed an inverted-U shape of age dependence. Compared to the cortical thickness, the normalized volume of subcortical regions exhibited more divergent trends, with some regions increasing, some decreasing, and some displaying inverted-U-shaped trends. The growth curve models for all brain regions demonstrated utilities in identifying brain atypicality. The percentile measures derived from the growth curves facilitate the identification of children with developmental speech and language disorders with an accuracy of 0.875 (area under the receiver operating characteristic curve: 0.943). Our results fill the knowledge gap in brain morphometrics in a critical development period and provide an avenue for individualized brain developmental status evaluation with demonstrated sensitivity. The brain growth charts are shared with the public (http://phi-group.top/resources.html).

NeurIPS Conference 2022 Conference Paper

SAViT: Structure-Aware Vision Transformer Pruning via Collaborative Optimization

Chuanyang Zheng
Zheyang Li
Kai Zhang
Zhi Yang
Wenming Tan
Jun Xiao
Ye Ren
Shiliang Pu

Vision Transformers (ViTs) yield impressive performance across various vision tasks. However, heavy computation and memory footprint make them inaccessible for edge devices. Previous works apply importance criteria determined independently by each individual component to prune ViTs. Considering that heterogeneous components in ViTs play distinct roles, these approaches lead to suboptimal performance. In this paper, we introduce joint importance, which integrates essential structural-aware interactions between components for the first time, to perform collaborative pruning. Based on the theoretical analysis, we construct a Taylor-based approximation to evaluate the joint importance. This guides pruning toward a more balanced reduction across all components. To further reduce the algorithm complexity, we incorporate the interactions into the optimization function under some mild assumptions. Moreover, the proposed method can be seamlessly applied to various tasks including object detection. Extensive experiments demonstrate the effectiveness of our method. Notably, the proposed approach outperforms the existing state-of-the-art approaches on ImageNet, increasing accuracy by 0. 7% over the DeiT-Base baseline while saving 50% FLOPs. On COCO, we are the first to show that 70% FLOPs of FasterRCNN with ViT backbone can be removed with only 0. 3% mAP drop. The code is available at https: //github. com/hikvision-research/SAViT.

YNIMG Journal 2021 Journal Article

Dynamic integration and segregation of amygdala subregional functional circuits linking to physiological arousal

Yimeng Zeng
Fuxiang Tao
Zaixu Cui
Liyun Wu
Jiahua Xu
Wenshan Dong
Chao Liu
Zhi Yang

The dynamical organization of brain networks is essential to support human cognition and emotion for rapid adaption to ever-changing environment. As the core nodes of emotion-related brain circuitry, the basolateral amygdala (BLA) and centromedial amygdala (CMA) as two major amygdalar nuclei, are recognized to play distinct roles in affective functions and internal states, via their unique connections with cortical and subcortical structures in rodents. However, little is known how the dynamical organization of emotion-related brain circuitry reflects internal autonomic responses in humans. Using resting-state functional magnetic resonance imaging (fMRI) with K-means clustering approach in a total of 79 young healthy individuals (cohort 1: 42; cohort 2: 37), we identified two distinct states of BLA- and CMA-based intrinsic connectivity patterns, with one state (integration) showing generally stronger BLA- and CMA-based intrinsic connectivity with multiple brain networks, while the other (segregation) exhibiting weaker yet dissociable connectivity patterns. In an independent cohort 2 of fMRI data with concurrent recording of skin conductance, we replicated two similar dynamic states and further found higher skin conductance level in the integration than segregation state. Moreover, machine learning-based Elastic-net regression analyses revealed that time-varying BLA and CMA intrinsic connectivity with distinct network configurations yield higher predictive values for spontaneous fluctuations of skin conductance level in the integration than segregation state. Our findings highlight dynamic functional organization of emotion-related amygdala nuclei circuits and networks and its links to spontaneous autonomic arousal in humans.

YNIMG Journal 2021 Journal Article

Impact of inter-individual variability on the estimation of default mode network in temporal concatenation group ICA

Yang Hu
Zhi Yang

Temporal concatenation group ICA (TC-GICA) is a widely used data-driven method to extract common functional brain networks among individuals. TC-GICA concatenates the time series of individual fMRI data and applies dimension reduction and ICA algorithms to decompose the data into group-level components. The default mode network (DMN) estimated using TC-GICA at relatively high model orders (i.e., large numbers of components) is split into multiple components. The split DMNs are topographically different from those estimated using other methods (e.g., seed-based correlation, clustering, graph theoretical analysis, and other ICA methods like gRAICAR and IVA-GL) and are inconsistent with the existing knowledge of DMN. We hypothesize that the "DMN-splitting'' phenomenon reflects the impact of inter-individual variability in data, which is propagated into the ICA decomposition via the data-concatenation step of TC-GICA. By systematically manipulating the amount of variability involved in the temporal concatenation in both simulated and several realistic datasets, we observed that as more variability was involved, the estimated DMN became less similar to the averaged functional connectivity (FC) pattern obtained using seed-based correlation analysis. The performance of the DMN estimation in TC-GICA also exhibited remarkable dependence on the model order settings. Further analyses revealed that the "DMN-splitting" in TC-GICA could be reproduced when involving large variability in the data-concatenation and performing ICA at high model orders. These results were replicated across multiple datasets and various software implementations. When applying ICA approaches that avoid temporal concatenation, such as gRAICAR and IVA-GL, to the same datasets, the estimated group-level DMN was more consistent with the seed-based FC pattern and was more robust to various model order settings. This study calls for caution when applying TC-GICA to datasets expected to have large inter-individual variability, such as pooling different experimental groups of subjects.

NeurIPS Conference 2021 Conference Paper

Node Dependent Local Smoothing for Scalable Graph Learning

Wentao Zhang
Mingyu Yang
Zeang Sheng
Yang Li
Wen Ouyang
Yangyu Tao
Zhi Yang
Bin Cui

Recent works reveal that feature or label smoothing lies at the core of Graph Neural Networks (GNNs). Concretely, they show feature smoothing combined with simple linear regression achieves comparable performance with the carefully designed GNNs, and a simple MLP model with label smoothing of its prediction can outperform the vanilla GCN. Though an interesting finding, smoothing has not been well understood, especially regarding how to control the extent of smoothness. Intuitively, too small or too large smoothing iterations may cause under-smoothing or over-smoothing and can lead to sub-optimal performance. Moreover, the extent of smoothness is node-specific, depending on its degree and local structure. To this end, we propose a novel algorithm called node-dependent local smoothing (NDLS), which aims to control the smoothness of every node by setting a node-specific smoothing iteration. Specifically, NDLS computes influence scores based on the adjacency matrix and selects the iteration number by setting a threshold on the scores. Once selected, the iteration number can be applied to both feature smoothing and label smoothing. Experimental results demonstrate that NDLS enjoys high accuracy -- state-of-the-art performance on node classifications tasks, flexibility -- can be incorporated with any models, scalability and efficiency -- can support large scale graphs with fast training.

NeurIPS Conference 2021 Conference Paper

RIM: Reliable Influence-based Active Learning on Graphs

Wentao Zhang
Yexin Wang
Zhenbang You
Meng Cao
Ping Huang
Jiulong Shan
Zhi Yang
Bin Cui

Message passing is the core of most graph models such as Graph Convolutional Network (GCN) and Label Propagation (LP), which usually require a large number of clean labeled data to smooth out the neighborhood over the graph. However, the labeling process can be tedious, costly, and error-prone in practice. In this paper, we propose to unify active learning (AL) and message passing towards minimizing labeling costs, e. g. , making use of few and unreliable labels that can be obtained cheaply. We make two contributions towards that end. First, we open up a perspective by drawing a connection between AL enforcing message passing and social influence maximization, ensuring that the selected samples effectively improve the model performance. Second, we propose an extension to the influence model that incorporates an explicit quality factor to model label noise. In this way, we derive a fundamentally new AL selection criterion for GCN and LP--reliable influence maximization (RIM)--by considering quantity and quality of influence simultaneously. Empirical studies on public datasets show that RIM significantly outperforms current AL methods in terms of accuracy and efficiency.

ICRA Conference 2021 Conference Paper

Smooth-RRT*: Asymptotically Optimal Motion Planning for Mobile Robots under Kinodynamic Constraints

Yiting Kang
Zhi Yang
Riya Zeng
Qi Wu

Nowadays, various algorithms based on the Rapidly-exploring Random Tree (RRT) methods are utilized to solve motion planning problems. Based on the RRT*, we developed a novel reconnection method that enables the planner to directly generate a smooth curved trajectory. Meanwhile, kinodynamic constraints of the robots are considered to generate the control input, which improves the feasibility of the algorithm. The trajectory planned by the Smooth-RRT* is significantly suitable for the non-holonomic robots. Planning tests are conducted in four scenarios to demonstrate performance of the proposed algorithm in comparison with the original RRT* and kinodynamic-RRT (Kino-RRT). Smooth-RRT* yields shorter and smoother planned path in all the scenarios compared with the Kino-RRT. It finds a solution with fewer expansion nodes than the RRT* under the same time consumption. The results demonstrate that the proposed algorithm can generate a smooth trajectory satisfied with the kinodynamic constraints and ensure the asymptotic optimality.

YNIMG Journal 2020 Journal Article

Individualized psychiatric imaging based on inter-subject neural synchronization in movie watching

Zhi Yang
Jinfeng Wu
Lihua Xu
Zhengzheng Deng
Yingying Tang
Jiaqi Gao
Yang Hu
Yiwen Zhang

The individual heterogeneity is a challenge to the prosperous promises of cutting-edge neuroimaging techniques for better diagnosis and early detection of psychiatric disorders. Individuals with similar clinical manifestations may result from very different pathophysiology. Conventional approaches based on comparing group-averages provide insufficient information to support the individualized diagnosis. Here we present an individualized imaging methodology that combines naturalistic imaging and the normative model. This paradigm adopts video clips with rich cognitive, social, and emotional contents to evoke synchronized brain dynamics of healthy participants and builds a spatiotemporal response norm. By comparing individual brain responses with the response norm, we could recognize patients using machine learning techniques. We applied this methodology to recognize first-episode drug-naïve schizophrenia patients in a dataset containing 72 patients and 54 healthy controls. Some segments of the video evoked more synchronized brain activity in the healthy controls than in the schizophrenia patients. We built a spatiotemporal response norm by averaging the brain responses of the healthy controls in a training set, and trained a classifier to recognize patients based on the differences between individual brain responses and the norm. The performance of the classifier was then evaluated using an independent test set. The mean accuracies from a 5-fold cross-validation were 0. 71–0. 78 depending on the parameters such as the number of features and the width of the sliding windows. These findings reflected the potential of this methodology towards a clinical tool for individualized diagnosis.

YNIMG Journal 2020 Journal Article

Reliability map of individual differences reflected in inter-subject correlation in naturalistic imaging

Jiaqi Gao
Gang Chen
Jinfeng Wu
YinShan Wang
Yang Hu
Ting Xu
Xi-Nian Zuo
Zhi Yang

Understanding individual differences in brain function is an essential aim of neuroscience. Naturalistic imaging links neural activity to real-life contexts and reflects individual differences in brain response. These unique features make it a promising tool for individualized psychiatry. An essential prerequisite for the extensive use of this paradigm is the reliable representation of inter-individual relationships. We used a test–retest approach to examine whether the naturalistic paradigm reliably represents inter-individual differences, which brain regions have the superior capability, and whether the ability alters with the contents of the stimuli. We quantified the reliability of the inter-subject relationships in repeated scans of two movie clips: a natural sight view and an emotion-evoking story. Besides statistical inference, we included resting-state scans, behavioral tests, and questionnaires as references for the comparison. The results showed that over one-third area of the brain could reliably characterize the inter-individual relationship, and the superior temporal lobe demonstrated comparable reliability representation with the State and Trait Anxiety Inventory. Furthermore, the temporal lobe regions could retain this capability across emotional movies with different contents. This study provides a base for pushing the naturalistic imaging paradigm towards clinical applications and proposes reliable target brain regions for future studies.

EAAI Journal 2019 Journal Article

Confidence-weighted safe semi-supervised clustering

Haitao Gan
Yingle Fan
Zhizeng Luo
Rui Huang
Zhi Yang

In this paper, we propose confidence-weighted safe semi-supervised clustering where prior knowledge is given in the form of class labels. In some applications, some samples may be wrongly labeled by the users. Therefore, our basic idea is that different samples should have different impacts or confidences on the clustering performance. In our algorithm, we firstly use unsupervised clustering to perform the dataset partition and compute the normalized confusion matrix N c. N c is used to estimate the safe confidence of each labeled sample based on the assumption that a correctly clustered sample should have a high confidence. Then we construct a local graph to model the relationship between the labeled and its nearest unlabeled samples through the clustering results. Finally, a confidence-weighted fidelity term and a graph-based regularization term are incorporated into the objective function of unsupervised clustering. In this case, on the one hand, the outputs of the labeled samples with high confidences are restricted to be the given prior labels. On the other hand, the outputs of the labeled ones with low confidences are forced to approach those of the local homogeneous unlabeled neighbors modeled by the local graph. Hence, the labeled samples are expected to be safely exploited which is the goal of safe semi-supervised clustering. To verify the effectiveness of our algorithm, we carry out some experiments over several datasets by comparison to the unsupervised and semi-supervised clustering methods and achieve the promising results.

NeurIPS Conference 2016 Conference Paper

A Bio-inspired Redundant Sensing Architecture

Anh Tuan Nguyen
Jian Xu
Zhi Yang

Sensing is the process of deriving signals from the environment that allows artificial systems to interact with the physical world. The Shannon theorem specifies the maximum rate at which information can be acquired. However, this upper bound is hard to achieve in many man-made systems. The biological visual systems, on the other hand, have highly efficient signal representation and processing mechanisms that allow precise sensing. In this work, we argue that redundancy is one of the critical characteristics for such superior performance. We show architectural advantages by utilizing redundant sensing, including correction of mismatch error and significant precision enhancement. For a proof-of-concept demonstration, we have designed a heuristic-based analog-to-digital converter - a zero-dimensional quantizer. Through Monte Carlo simulation with the error probabilistic distribution as a priori, the performance approaching the Shannon limit is feasible. In actual measurements without knowing the error distribution, we observe at least 2-bit extra precision. The results may also help explain biological processes including the dominance of binocular vision, the functional roles of the fixational eye movements, and the structural mechanisms allowing hyperacuity.

JBHI Journal 2016 Journal Article

A Wavelet-Based Artifact Reduction From Scalp EEG for Epileptic Seizure Detection

Md Kafiul Islam
Amir Rastegarnia
Zhi Yang

This paper presents a method to reduce artifacts from scalp EEG recordings to facilitate seizure diagnosis/detection for epilepsy patients. The proposed method is primarily based on stationary wavelet transform and takes the spectral band of seizure activities (i. e. , 0. 5–29 Hz) into account to separate artifacts from seizures. Different artifact templates have been simulated to mimic the most commonly appeared artifacts in real EEG recordings. The algorithm is applied on three sets of synthesized data including fully simulated, semi-simulated, and real data to evaluate both the artifact removal performance and seizure detection performance. The EEG features responsible for the detection of seizures from nonseizure epochs have been found to be easily distinguishable after artifacts are removed, and consequently, the false alarms in seizure detection are reduced. Results from an extensive experiment with these datasets prove the efficacy of the proposed algorithm, which makes it possible to use it for artifact removal in epilepsy diagnosis as well as other applications regarding neuroscience studies.

YNIMG Journal 2015 Journal Article

Cortisol awakening response predicts intrinsic functional connectivity of the medial prefrontal cortex in the afternoon of the same day

Jianhui Wu
Shen Zhang
Wanqing Li
Shaozheng Qin
Yong He
Zhi Yang
Tony W. Buchanan
Chao Liu

Cortisol awakening response (CAR) is the cortisol secretory activity in the first 30–60min immediately after awakening in the morning. Alterations in CAR as a trait have been associated with changes in the brain structure and function. CAR also fluctuates over days. Little, however, is known about the relationship between CAR as a state and brain activity. Using resting-state functional magnetic resonance imaging (fMRI), we investigated whether the CAR predicts intrinsic functional connectivity (FC) of the brain in the afternoon of the same day. Data from forty-nine healthy participants were analyzed. Salivary cortisol levels were assessed immediately after awakening and 15, 30 and 60min after awakening, and resting-state fMRI data were obtained in the afternoon. Global FC strength (FCS) of each voxel was computed to provide a whole-brain characterization of intrinsic functional architecture. Correlation analysis was used to examine whether CAR predicts the intrinsic FC of core brain networks. We observed that the CAR was positively correlated with the FCS of the medial prefrontal cortex (mPFC). Further analysis revealed that higher CAR predicted stronger positive mPFC connectivity with regions in the default mode network. Our findings suggest that the HPA activity after awakening in the early morning may predict intrinsic functional connectivity of mPFC at rest in the afternoon of the same day.

YNIMG Journal 2014 Journal Article

Connectivity trajectory across lifespan differentiates the precuneus from the default network

Zhi Yang
Catie Chang
Ting Xu
Lili Jiang
Daniel A. Handwerker
F. Xavier Castellanos
Michael P. Milham
Peter A. Bandettini

The default network of the human brain has drawn much attention due to its relevance to various brain disorders, cognition, and behavior. However, its functional components and boundaries have not been precisely defined. There is no consensus as to whether the precuneus, a hub in the functional connectome, acts as part of the default network. This discrepancy is more critical for brain development and aging studies: it is not clear whether age has a stronger impact on the default network or precuneus, or both. We used Generalized Ranking and Averaging Independent Component Analysis by Reproducibility (gRAICAR) to investigate the lifespan trajectories of intrinsic functional networks. By estimating individual-specific spatial components and aligning them across subjects, gRAICAR measures the spatial variation of component maps across a population without constraining the same components to appear in every subject. In a cross-lifespan fMRI dataset (N=126, 7–85years old), we observed stronger age dependence in the spatial pattern of a precuneus–dorsal posterior cingulate cortex network compared to the default network, despite the fact that the two networks exhibit considerable spatial overlap and temporal correlation. These results remained even when analyses were restricted to a subpopulation with very similar head motion across age. Our analyses further showed that the two networks tend to merge with increasing age. Post-hoc analyses of functional connectivity confirmed the distinguishable cross-lifespan trajectories between the two networks. Based on these observations, we proposed a dynamic model of cross-lifespan functional segregation and integration between the two networks, suggesting that the precuneus network may have a different functional role than the default network, which declines with age. These findings have implications for understanding the functional roles of the default network, gaining insight into its dynamics throughout life, and guiding interpretation of alterations in brain disorders.

YNIMG Journal 2014 Journal Article

Using fMRI to decode true thoughts independent of intention to conceal

Zhi Yang
Zirui Huang
Javier Gonzalez-Castillo
Rui Dai
Georg Northoff
Peter Bandettini

Multi-variate pattern analysis (MVPA) applied to BOLD-fMRI has proven successful at decoding complicated fMRI signal patterns associated with a variety of cognitive processes. One cognitive process, not yet investigated, is the mental representation of “Yes/No” thoughts that precede the actual overt response to a binary “Yes/No” question. In this study, we focus on examining: (1) whether spatial patterns of the hemodynamic response carry sufficient information to allow reliable decoding of “Yes/No” thoughts; and (2) whether decoding of “Yes/No” thoughts is independent of the intention to respond honestly or dishonestly. To achieve this goal, we conducted two separate experiments. Experiment 1, collected on a 3T scanner, examined the whole brain to identify regions that carry sufficient information to permit significantly above-chance prediction of “Yes/No” thoughts at the group level. In Experiment 2, collected on a 7T scanner, we focused on the regions identified in Experiment 1 to examine the capability of achieving high decoding accuracy at the single subject level. A set of regions – namely right superior temporal gyrus, left supra-marginal gyrus, and left middle frontal gyrus – exhibited high decoding power. Decoding accuracy for these regions increased with trial averaging. When 18 trials were averaged, the median accuracies were 82. 5%, 77. 5%, and 79. 5%, respectively. When trials were separated according to deceptive intentions (set via experimental cues), and classifiers were trained on honest trials, but tested on trials where subjects were asked to deceive, the median accuracies of these regions still reached 66%, 75%, and 78. 5%. These results provide evidence that concealed “Yes/No” thoughts are encoded in the BOLD signal, retaining some level of independence from the subject’s intentions to answer honestly or dishonestly. These findings also suggest the theoretical possibility for more efficient brain-computer interfaces where subjects only need to think their answers to communicate.

YNIMG Journal 2013 Journal Article

Toward reliable characterization of functional homogeneity in the human brain: Preprocessing, scan duration, imaging resolution and computational space

Xi-Nian Zuo
Ting Xu
Lili Jiang
Zhi Yang
Xiao-Yan Cao
Yong He
Yu-Feng Zang
F. Xavier Castellanos

While researchers have extensively characterized functional connectivity between brain regions, the characterization of functional homogeneity within a region of the brain connectome is in early stages of development. Several functional homogeneity measures were proposed previously, among which regional homogeneity (ReHo) was most widely used as a measure to characterize functional homogeneity of resting state fMRI (R-fMRI) signals within a small region (Zang et al. , 2004). Despite a burgeoning literature on ReHo in the field of neuroimaging brain disorders, its test–retest (TRT) reliability remains unestablished. Using two sets of public R-fMRI TRT data, we systematically evaluated the ReHo's TRT reliability and further investigated the various factors influencing its reliability and found: 1) nuisance (head motion, white matter, and cerebrospinal fluid) correction of R-fMRI time series can significantly improve the TRT reliability of ReHo while additional removal of global brain signal reduces its reliability, 2) spatial smoothing of R-fMRI time series artificially enhances ReHo intensity and influences its reliability, 3) surface-based R-fMRI computation largely improves the TRT reliability of ReHo, 4) a scan duration of 5min can achieve reliable estimates of ReHo, and 5) fast sampling rates of R-fMRI dramatically increase the reliability of ReHo. Inspired by these findings and seeking a highly reliable approach to exploratory analysis of the human functional connectome, we established an R-fMRI pipeline to conduct ReHo computations in both 3-dimensions (volume) and 2-dimensions (surface).

YNIMG Journal 2012 Journal Article

Generalized RAICAR: Discover homogeneous subject (sub)groups by reproducibility of their intrinsic connectivity networks

Zhi Yang
Xi-Nian Zuo
Peipei Wang
Zhihao Li
Stephen M. LaConte
Peter A. Bandettini
Xiaoping P. Hu

Existing spatial independent component analysis (ICA) methods for multi-subject fMRI datasets have mainly focused on detecting common components across subjects, under the assumption that all the subjects in a group share the same (identical) components. However, as a data-driven approach, ICA could potentially serve as an exploratory tool at multi-subject level, and help us uncover inter-subject differences in patterns of connectivity (e. g. , find subtypes in patient populations). In this work, we propose a methodology named gRAICAR that exploits the data-driven nature of ICA to allow discovery of sub-groupings of subjects based on reproducibility of their ICA components. This technique allows us not only to find highly reproducible common components across subjects but also to explore (without a priori subject groupings) components that could classify all subjects into sub-groups. gRAICAR generalizes the reproducibility framework previously developed for single subjects (Ranking and averaging independent component analysis by reproducibility—RAICAR—Yang et al. , Hum Brain Mapp, 2008) to multiple-subject analysis. For each group-level component, gRAICAR generates its reproducibility matrix and further computes two metrics, inter-subject consistency and intra-subject reliability, to characterize inter-subject variability and reflect contributions from individual subjects. Nonparametric tests are employed to examine the significance of both the inter-subject consistency and the separation of subject groups reflected in the component. Our validations based on simulated and experimental resting-state fMRI datasets demonstrated the advantage of gRAICAR in extracting features reflecting potential subject groupings. It may facilitate discovery of the underlying brain functional networks with substantial potential to inform our understandings of development, neurodegenerative conditions, and psychiatric disorders.

NeurIPS Conference 2009 Conference Paper

Noise Characterization, Modeling, and Reduction for In Vivo Neural Recording

Zhi Yang
Qi Zhao
Edward Keefer
Wentai Liu

Studying signal and noise properties of recorded neural data is critical in developing more efficient algorithms to recover the encoded information. Important issues exist in this research including the variant spectrum spans of neural spikes that make it difficult to choose a global optimal bandpass filter. Also, multiple sources produce aggregated noise that deviates from the conventional white Gaussian noise. In this work, the spectrum variability of spikes is addressed, based on which the concept of adaptive bandpass filter that fits the spectrum of individual spikes is proposed. Multiple noise sources have been studied through analytical models as well as empirical measurements. The dominant noise source is identified as neuron noise followed by interface noise of the electrode. This suggests that major efforts to reduce noise from electronics are not well spent. The measured noise from in vivo experiments shows a family of 1/f^{x} (x=1. 5\pm 0. 5) spectrum that can be reduced using noise shaping techniques. In summary, the methods of adaptive bandpass filtering and noise shaping together result in several dB signal-to-noise ratio (SNR) enhancement.

NeurIPS Conference 2008 Conference Paper

Spike Feature Extraction Using Informative Samples

Zhi Yang
Qi Zhao
Wentai Liu

This paper presents a spike feature extraction algorithm that targets real-time spike sorting and facilitates miniaturized microchip implementation. The proposed algorithm has been evaluated on synthesized waveforms and experimentally recorded sequences. When compared with many spike sorting approaches our algorithm demonstrates improved speed, accuracy and allows unsupervised execution. A preliminary hardware implementation has been realized using an integrated microchip interfaced with a personal computer.