A5100449095- Functional Brain Connectivity Studies
- Advanced MRI Techniques and Applications
- Cerebrovascular and Carotid Artery Diseases
- Advanced Neuroimaging Techniques and Applications
- Intracranial Aneurysms: Treatment and Complications
- Mental Health Research Topics
- EEG and Brain-Computer Interfaces
- Medical Image Segmentation Techniques
- Sparse and Compressive Sensing Techniques
- Recommender Systems and Techniques
- Neurological disorders and treatments
- Parkinson's Disease Mechanisms and Treatments
- Advanced Graph Neural Networks
- Brain Tumor Detection and Classification
- Neural dynamics and brain function
- Text and Document Classification Technologies
- Hearing Loss and Rehabilitation
- Advanced Neural Network Applications
- Industrial Vision Systems and Defect Detection
- Distributed Control Multi-Agent Systems
- Neural Networks and Reservoir Computing
- Technical Engine Diagnostics and Monitoring
- Imbalanced Data Classification Techniques
- Cultural Differences and Values
- Tensor decomposition and applications
Aviation General Hospital
2024
Beijing Institute of Technology
2010-2024
Beihang University
2024
Shanghai Center for Brain Science and Brain-Inspired Technology
2024
Shanghai Institute for Science of Science
2024
Fudan University
2024
University of Utah
2024
Tiangong University
2023-2024
University of Washington
2024
Tsinghua University
2019
Alzheimer's disease (AD) is one of the most common progressive and irreversible neurodegenerative diseases. The study pathological mechanism AD early-stage diagnosis essential important. Subjective cognitive decline (SCD), first at-risk stage occurring prior to amnestic mild impairment (aMCI), great research value has gained our interest. To investigate entire development pathology efficiently, we proposed a machine learning classification method based on multimodal support vector (SVM)...
Representation learning in heterogeneous graphs with massive unlabeled data has aroused great interest. The heterogeneity of not only contains rich information, but also raises difficult barriers to designing unsupervised or self-supervised (SSL) strategies. Existing methods such as random walk-based approaches are mainly dependent on the proximity information neighbors and lack ability integrate node features into a higher-level representation. Furthermore, previous frameworks usually...
Self-construal (orientations of independence and interdependence) is a fundamental concept that guides human behaviour, it linked to large number brain regions. However, understanding the connectivity these regions critical principles underlying self-functions are lacking. Because activity self-related processes intrinsic, resting-state method has received substantial attention. Here, we focused on functional matrices based asymmetry as indexed by differential partition located in mirrored...
Subthalamic nucleus deep brain stimulation (STN-DBS) is an effective invasive treatment for advanced Parkinson's disease (PD) at present. Due to the invasiveness and cost of operations, a reliable tool required predict outcome therapy in clinical decision-making process. This work aims investigate whether topological network functional connectivity states can DBS without medication. 50 patients were recruited extract features related improvement rate PD after STN-DBS train machine learning...
The current two-stage detectors remarkably benefit from hybrid representation of points and 3-D voxels, but they have high time cost leave room for improving the accuracy small objects. On contrary, 2-D voxel-based methods tend to good efficiency better performance An intuitive idea optimizing a algorithm is use backbone. However, naive substitution cannot achieve optimal joint learning each may cause decrease in accuracy. In this article, we propose point–voxel RCNN (HPV-RCNN), novel point...
Deep learning-based label noise learning methods provide promising solutions for hyperspectral image (HSI) classification with noisy labels. Currently, based on deep improve their performance by modifying one aspect, such as designing a robust loss function, revamping the network structure, or adding adaptation layer. However, these face difficulties in coping relatively high situations. To address this issue, paper proposes unified framework dual-network structure. The goal is to enhance...
The circle of Willis (COW) is a crucial mechanism for cerebral collateral circulation. This proof-of-concept study aims to develop and assess an analysis method characterize the hemodynamics arterial segments in COW by using spin-labeling (ASL) based non-contrast-enhanced dynamic MR angiography (dMRA).
Acquiring reviewers for academic submissions is a challenging recommendation scenario. Recent graph learning-driven models have made remarkable progress in the field of recommendation, but their performance reviewer task may suffer from significant false negative issue. This arises assumption that unobserved edges represent samples. In fact, mechanism anonymous review results inadequate exposure interactions between and submissions, leading to higher number compared those caused by declining...
Parkinson's disease (PD) is a multi-systemic in the brain arising from dysfunction of several neural networks. The diagnosis and treatment PD have gained more attention for clinical researchers. While there been many fMRI studies about functional topological changes patients, whether dynamic connectivity can predict drug therapy effect still unclear. primary objective this study was to assess large-scale efficiency network are detectable explore severity level (UPDRS-III) after be predicted...
The self-construal is one of the most significant cultural markers in humans. Accordingly, mapping relationship between brain activity and contributes to understanding nature such psychological traits. Existing studies have mainly focused on static functional activities specific regions. However, evidence has suggested that connectivity (FC) network dynamic over time high-level processes might require collaboration among multiple In present study, we explored connection patterns two...
Tracking stress-induced brain activity and connectivity dynamically examining activity/connectivity-associated recovery ability after stress might be an effective way of detecting vulnerability.Using two widely used paradigms, a speech task (social stress) mathematical calculation (mental loading stress), we examined common changes in regional homogeneity (ReHo) functional (FC) before, during, the stressful tasks thirty-nine college students. A counting breath relaxation was employed as...
Levodopa is the most-commonly used therapy for Parkinson's Disease (PD). Imaging findings show increased cerebral blood flow (CBF) response to levodopa, but artery morphological change less studied.To investigate effect of levodopa on arteries and CBF.Prospective.57 PD patients (56 ± 10 years, 26 males) 17 age-matched healthy controls (AMC, 57 9 were scanned at baseline (OFF). Patients rescanned 50 minutes after taking (ON).3 T; Simultaneous noncontrast angiography intraplaque imaging (SNAP)...
Parkinson's disease (PD) is a neurodegenerative that associated with motor and non-motor symptoms caused by lack of dopamine in the substantia nigra brain. Subthalamic nucleus deep brain stimulation (STN-DBS) widely accepted therapy PD mainly inserts electrodes into both sides The effect STN-DBS was for function, so this study focused on recovery function after DBS. Hemispherical asymmetry network considered to be potential indicator diagnosing patients. This investigated value hemispheric...