A5100459489- Advanced Neural Network Applications
- Privacy-Preserving Technologies in Data
- Medical Image Segmentation Techniques
- Distributed and Parallel Computing Systems
- Cloud Computing and Resource Management
- Advanced Image and Video Retrieval Techniques
- Medical Imaging and Analysis
- Advanced Computational Techniques and Applications
- Functional Brain Connectivity Studies
- EEG and Brain-Computer Interfaces
- Brain Tumor Detection and Classification
- Neural dynamics and brain function
- Atomic and Subatomic Physics Research
- Adversarial Robustness in Machine Learning
- Advanced Vision and Imaging
- Advanced MRI Techniques and Applications
- Video Surveillance and Tracking Methods
- Advanced Malware Detection Techniques
- Cognitive Radio Networks and Spectrum Sensing
- Traumatic Brain Injury Research
- IoT and Edge/Fog Computing
- Traffic control and management
- Advanced MIMO Systems Optimization
- Cryptography and Data Security
- Distributed Control Multi-Agent Systems
Shanghai Jiao Tong University
2015-2024
Beijing Institute of Technology
2023-2024
Dalian Medical University
2024
Xiamen University
2011-2024
China Jiliang University
2022-2024
Chinese Academy of Sciences
2018-2024
Xi'an High Tech University
2022-2024
Yalong Hydro (China)
2024
Ningbo Institute of Industrial Technology
2023-2024
Chongqing University of Technology
2021-2024
Abstract A strategy for using tissue water as a concentration standard in 1 H magnetic resonance spectroscopic imaging studies on the brain is presented, and potential errors that may arise when method used are examined. The sensitivity of to estimates different compartment relaxation times shown be small at short echo (TEs). Using data from healthy human subjects, it image segmentation approaches commonly account partial volume effects (SPM2, FSL's FAST, K‐means) lead metabolite levels,...
Deep learning-based semi-supervised learning (SSL) algorithms have led to promising results in medical images segmentation and can alleviate doctors' expensive annotations by leveraging unlabeled data. However, most of the existing SSL literature tend regularize model training perturbing networks and/or Observing that multi/dual-task attends various levels information which inherent prediction perturbation, we ask question this work: explicitly build task-level regularization rather than...
A key challenge in federated learning (FL) is the statistical heterogeneity that impairs generalization of global model on each client. To address this, we propose a method Federated with Adaptive Local Aggregation (FedALA) by capturing desired information for client models personalized FL. The component FedALA an (ALA) module, which can adaptively aggregate downloaded and local towards objective to initialize before training iteration. evaluate effectiveness FedALA, conduct extensive...
Traumatic brain injury (TBI) is a leading cause of sustained impairment in military and civilian populations. However, mild (and some moderate) TBI can be difficult to diagnose due lack obvious external injuries because the are often not visible on conventional acute MRI or CT. Injured tissues patients generate pathological low-frequency neuronal magnetic signal (delta waves 1–4 Hz) that measured localized by magnetoencephalography (MEG). We hypothesize abnormal MEG delta originate from gray...
Video object detection is more challenging than image because of the deteriorated frame quality. To enhance feature representation, state-of-the-art methods propagate temporal information into by aligning and aggregating entire maps from multiple nearby frames. However, restricted map's low storage-efficiency vulnerable content-address allocation, long-term not fully stressed these methods. In this work, we propose first guided external memory network for online video detection....
With the rapid growth of video data, summarization technique plays a key role in reducing people's efforts to explore content videos by generating concise but informative summaries. Though supervised approaches have been well studied and achieved state-of-the-art performance, unsupervised methods are still highly demanded due intrinsic difficulty obtaining high-quality annotations. In this paper, we propose novel yet simple method with attentive conditional Generative Adversarial Networks...
Recently, personalized federated learning (pFL) has attracted increasing attention in privacy protection, collaborative learning, and tackling statistical heterogeneity among clients, e.g., hospitals, mobile smartphones, etc. Most existing pFL methods focus on exploiting the global information client-level model parameters while neglecting that data is source of these two kinds information. To address this, we propose Federated Conditional Policy (FedCP) method, which generates a conditional...
Vehicular Ad-hoc Network (VANET) is a new application of Mobile (MANET) in the field Inter-vehicle communication. As high mobility vehicles, some traditional MANET routing protocols may not fit VANET. In this paper, we propose cluster-based directional protocol (CBDRP) for highway scenarios, which header cluster selects another according to moving direction vehicle forward packets. Simulation results shows CBDRP can solve problem link stability VANET, realizing reliable and rapid data transmission.
A modified probabilistic neural network (PNN) for brain tissue segmentation with magnetic resonance imaging (MRI) is proposed. In this approach, covariance matrices are used to replace the singular smoothing factor in PNN's kernel function, and weighting factors added pattern of summation layer. This weighted (WPNN) classifier can account partial volume effects, which exist commonly MRI, not only final result stage, but also modeling process. It adopts self-organizing map (SOM) overly...
Traumatic brain injury (TBI) is a leading cause of sustained impairment in military and civilian populations. However, mild TBI (mTBI) can be difficult to detect using conventional MRI or CT. Injured tissues mTBI patients generate abnormal slow-waves (1–4 Hz) that measured localized by resting-state magnetoencephalography (MEG). In this study, we develop voxel-based whole-brain MEG slow-wave imaging approach for detecting abnormality with on single-subject basis. A normative database source...
Post-traumatic stress disorder (PTSD) is a leading cause of sustained impairment, distress, and poor quality life in military personnel, veterans, civilians. Indirect functional neuroimaging studies using PET or fMRI with fear-related stimuli support PTSD neurocircuitry model that includes amygdala, hippocampus, ventromedial prefrontal cortex (vmPFC). However, it not clear if this can fully account for abnormalities detected directly by electromagnetic-based source imaging techniques...
Transition metal-catalyzed remote hydrofunctionalization of alkenes is an efficient method to realize C(sp3)–H functionalization. However, with unactivated amines and alcohols has not been successfully developed date. Herein, we report nickel-catalyzed hydroamination hydroetherification alcohols, accessing a series gem-diamine N,O-acetal derivatives in good high yields (up 93%) exclusive regioselectivities. The mechanistic investigations DFT computations indicated that the use...
Federated Learning (FL) is popular for its privacy-preserving and collaborative learning capabilities. Recently, personalized FL (pFL) has received attention ability to address statistical heterogeneity achieve personalization in FL. However, from the perspective of feature extraction, most existing pFL methods only focus on extracting global or information during local training, which fails meet goals pFL. To this, we propose a new method, named GPFL, simultaneously learn each client. We...
Blast mild traumatic brain injury (mTBI) is a leading cause of sustained impairment in military service members and veterans. However, the mechanism persistent disability not fully understood. The present study investigated disturbances functioning mTBI participants using source-imaging-based approach to analyze functional connectivity (FC) from resting-state magnetoencephalography (rs-MEG). Study included 26 active-duty or veterans who had blast with post-concussive symptoms, 22 healthy...
With the popularity of machine learning on many applications, data privacy has become a severe issue when is applied in real world. Federated (FL), an emerging paradigm learning, aims to train centralized model while distributing training among large number clients order avoid leaking, which attracted great attention recently. However, distributed scheme FL susceptible different kinds attacks. Existing defense systems mainly utilize weight analysis identify malicious with limitations. For...
Graphene oxide modulated dual S-scheme ultrathin heterojunctions with iron phthalocyanine and phase-mixed bismuth molybdate as wide visible-light catalysts for tetracycline degradation.
With the continuous innovation of physical education methods, Comprehension Teaching Method has been effectively promoted, primarily used in teaching modern ball games, focusing on reflecting characteristics and tactics games. As an essential part youth sports China, applying to children's basketball can better enhance students' enthusiasm, improve their overall cognition, strengthen adaptability. This paper employs methods such as literature review logical analysis, combined with...
Although impairments related to somatosensory perception are common in schizophrenia, they have rarely been examined functional imaging studies. In the present study, magnetoencephalography (MEG) was used identify neural networks that support attention stimuli healthy adults and abnormalities these patient with schizophrenia. A median-nerve oddball task probe stimuli, an advanced, high-resolution MEG source-imaging method applied assess activity throughout brain. nineteen subjects,...