Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks284 from public data
- Multimodal classification of Alzheimer's disease and mild cognitive impairment1,315
This paper proposes to combine three modalities of biomarkers, i.e., MRI, FDG-PET, and CSF biomarkers, to discriminate between AD (or MCI) and healthy controls, using a kernel combination method, and shows considerably better performance compared to the case of using an individual modality of biomarkers.
- 484
- 3D conditional generative adversarial networks for high-quality PET image estimation at low dose449
Experimental results show that the proposed 3D c‐GANs method outperforms the benchmark methods and achieves much better performance than the state‐of‐the‐art methods in both qualitative and quantitative measures.
- 298
- METransformer: Radiology Report Generation by Transformer with Multiple Learnable Expert Tokens232
The proposed METransformer, a method to realize a “multi-expert joint diagnosis” mechanism to upgrade the existing “single expert” framework commonly seen in the current literature, enjoys the merits of an ensemble-based approach but through a manner that is computationally more efficient and supports more sophisticated interactions among experts.
- HEp-2 Cell Image Classification With Deep Convolutional Neural Networks226
The results on benchmark datasets demonstrate that the proposed framework can effectively outperform existing models by properly applying data augmentation, and the CNN-based framework has excellent adaptability across different datasets, which is highly desirable for cell image classification under varying laboratory settings.
- Discriminant analysis of longitudinal cortical thickness changes in Alzheimer's disease using dynamic and network features210
The decreasing of network clustering coefficient of the olfactory cortex region was found in the AD patients, which suggests the olfactory dysfunction, and a significant decrease of the network clustering coefficient is found in the Progressive-MCI group, which indicates the degenerated wiring efficiency of the brain network due to AD.
- 3D Auto-Context-Based Locality Adaptive Multi-Modality GANs for PET Synthesis205
Experimental results show that the proposed 3D auto-context-based locality adaptive multi-modality generative adversarial networks model (LA-GANs) outperforms the traditional multi- modality fusion methods used in deep networks, as well as the state-of-the-art PET estimation approaches.
- Semi-supervised medical image segmentation via a tripled-uncertainty guided mean teacher model with contrastive learning204
This paper introduces two auxiliary tasks, i.e., a foreground and background reconstruction task for capturing semantic information and a signed distance field prediction task for imposing shape constraint, and proposes an uncertainty weighted integration (UWI) strategy for yielding the segmentation predictions of the teacher.
- 191
- Cross-sectional and Longitudinal Analysis of the Relationship Between Aβ Deposition, Cortical Thickness, and Memory in Cognitively Unimpaired Individuals and in Alzheimer Disease187
The notion that Aβ deposition is not a benign process and that interventions with anti-Aβ therapy at these early stages have a higher chance to be effective is supported.
- A Survey on Efficient Vision Transformers: Algorithms, Techniques, and Performance Benchmarking177
The strategies used to make Vision Transformer efficient are mathematically defined, described and discusses state-of-the-art methodologies, and their performances over different application scenarios are analyzed.
- A Novel Unsupervised Camera-Aware Domain Adaptation Framework for Person Re-Identification166
This paper highlights the presence of camera-level sub-domains as a unique characteristic in person Re-ID, and develops a “camera-aware” domain adaptation method via adversarial learning, and exploits the temporal continuity in each camera of target domain to create discriminative information.
- Augmentation Matters: A Simple-Yet-Effective Approach to Semi-Supervised Semantic Segmentation156
This work proposes AugSeg, a simple and clean approach that focuses mainly on data perturbations to boost the SSS performance, and adopts a simplified intensity-based augmentation that selects a random number of data transformations with uniformly sampling distortion strengths from a continuous space.
- 129
- Multiple Kernel Learning in the Primal for Multimodal Alzheimer’s Disease Classification129
This paper proposes a novel multiple kernel-learning framework to combine multimodal features for AD classification, which is scalable and easy to implement, and imposes the mixed L21 norm constraint on the kernel weights to enforce group sparsity among different feature modalities.
- 117
- Adaptive rectification based adversarial network with spectrum constraint for high-quality PET image synthesis113
An adaptive rectification based generative adversarial network with spectrum constraint, named AR-GAN, which uses low-dose PET (LPET) image to synthesize standard-dose SPET image of high-quality and outperform other state-of-the-art image synthesis approaches.
- 102
- Conflict-Based Cross-View Consistency for Semi-Supervised Semantic Segmentation100
A new conflict-based cross-view consistency (CCVC) method based on a two-branch co-training framework which aims at enforcing the two sub-nets to learn informative features from irrelevant views and validate the new CCVC approach on the SSS benchmark datasets where the method achieves new state-of-the-art performance.
- Epileptic Seizure Forecasting With Generative Adversarial Networks100
This article proposes an approach that can make use of not only labeled EEG signals but also the unlabeled ones which are more accessible, using the short-time Fourier transform on 28-s EEG windows as a pre-processing step and is the first application of GAN to seizure forecasting.
- Beyond Covariance: Feature Representation with Nonlinear Kernel Matrices100
This paper proposes an open framework to use the kernel matrix over feature dimensions as a generic representation and discusses its properties and advantages, which significantly elevates covariance representation to the unlimited opportunities provided by this new representation.
- ASMFS: Adaptive-similarity-based multi-modality feature selection for classification of Alzheimer's disease94
A novel multi-modality feature selection method, which performs feature selection and local similarity learning simultaniously, and a similarity matrix is learned by jointly considering different imaging modalities is proposed.
- 88
- 84
- Epileptic Seizure Classification With Symmetric and Hybrid Bilinear Models84
This paper proposes a novel hybrid bilinear deep learning network with an application in the clinical procedures of epilepsy classification diagnosis, where the use of surface electroencephalogram (sEEG) and audiovisual monitoring is standard practice.
- Hierarchical Anatomical Brain Networks for MCI Prediction: Revisiting Volumetric Measures84
The proposed approach improves the accuracy of MCI prediction from (of conventional volumetric features) to ( of hierarchical network features), evaluated using data sets randomly drawn from the ADNI (Alzheimer's Disease Neuroimaging Initiative) dataset.
- 83
- Medical Image Synthesis via Deep Learning82
This chapter will focus on introducing typical CNNs and GANs models for medical image synthesis, and elaborate the recent work about low-dose to high-dose PET image synthesisation, and cross-modality MR images synthesis, using these models.
- 77
- 74
- 72
- Single- and Multiple-Dose Trials to Determine the Pharmacokinetics, Safety, Tolerability, and Sex Effect of Oral Ginsenoside Compound K in Healthy Chinese Volunteers69
CK was safe and well-tolerated over the treatment period, and sex differences were characterized by a higher exposure in females than males with the single administration after breakfast, and no severe AEs were observed.
- 68
- MR-Less Surface-Based Amyloid Assessment Based on 11C PiB PET64
An MR-less Aβ imaging quantification method that requires only PET images for reliable Aβ burden estimations is developed and demonstrated accuracy suggests that the proposed method could be an effective clinical inspection tool for A β imaging scans when MRI images are unavailable.
- 63
- DA-DSUnet: Dual Attention-based Dense SU-net for automatic head-and-neck tumor segmentation in MRI images62
A Dual Attention-based Dense SU-net (DA-DSUnet) framework for automatic NPC segmentation is proposed, an encoder-decoder network taking 2D NPC MRI slices as input and outputting the corresponding segmentation results.
- Unsupervised brain tumor segmentation using a symmetric-driven adversarial network62
It is demonstrated that symmetric features presenting variations can be modelled using unannotated normal MR images and thus be used in segmenting tumors and thus be used in segmenting tumors.
- Changes in hippocampal synapses and learning-memory abilities in a streptozotocin-treated rat model and intervention by using fasudil hydrochloride62
The results indicate that STZ induced deficit in learning/memory, decrease in SYP expression, degeneration in synaptic structures, and increase in the expressions of p-LIMK2 and p-cofilin were reversed by the administration of FH, suggesting that FH has anti-dementia properties that protect synaptic structure and function.
- A robust and efficient AI assistant for breast tumor segmentation from DCE-MRI via a spatial-temporal framework59
An artificial intelligence (AI) assistant to automatically segment breast tumors by capturing dynamic changes in multi-phase DCE-MRI with a spatial-temporal framework is developed and will promote the application of AI in more clinical diagnostic practices regarding breast cancer.
- Instance-Specific and Model-Adaptive Supervision for Semi-Supervised Semantic Segmentation58
iMAS learns from unlabeled instances progressively by weighing their corresponding consistency losses based on the evaluated hardness and dynamically adjusts the augmentation for each instance such that the distortion degree of augmented instances is adapted to the model's generalization capability across the training course.
- MutexMatch: Semi-Supervised Learning With Mutex-Based Consistency Regularization58
This article aims to utilize low-confidence samples in a novel way with the proposed mutex-based consistency regularization, namely MutexMatch, which achieves superior performance on multiple benchmark datasets, i.e., Canadian Institute for Advanced Research (CIFAR)-10, CIFAR-100, street view house numbers, self-taught learning 10, and mini-ImageNet.
- GPT4Video: A Unified Multimodal Large Language Model for lnstruction-Followed Understanding and Safety-Aware Generation57
This paper presents GPT4Video, a unified framework that seamlessly and lightly integrates with LLMs, visual feature extractors, and stable diffusion generative models for cohesive video understanding and generation, and explores a text-only finetuning approach to equip models for instruction-following and safeguarding in multimodal conversations.
- BriNet: Towards Bridging the Intra-class and Inter-class Gaps in One-Shot Segmentation56
A framework to bridge the gap between information interaction between query and support images and to precisely localize the query objects is proposed, which outperforms other competitive methods and leads to a new state-of-the-art on both PASCAL VOC and MSCOCO dataset.
- Dense Video Captioning Using Graph-Based Sentence Summarization56
A graph-based partition-and-summarization (GPaS) framework for dense video captioning within two stages that effectively exploits the relationship between semantic words for summarization and achieves this goal by treating semantic words as the nodes in a graph.
- 55
- LaSSL: Label-Guided Self-Training for Semi-supervised Learning53
This paper proposes a Label-guided Self-training approach to Semi-supervised Learning (LaSSL), which improves pseudo-label generations from two mutually boosted strategies and evaluates LaSSL on several classification benchmarks under partially labeled settings and demonstrates its superiority over the state-of-the-art approaches.
- A Comprehensive Study of GPT-4V’s Multimodal Capabilities in Medical Imaging51
This paper presents a comprehensive evaluation of GPT-4V's capabilities across diverse medical imaging tasks, including Radiology Report Generation, Medical Visual Question Answering (VQA), and Visual Grounding, and finds the limitations of conventional evaluation metrics like the BLEU score.
- 50
- 3D multi-modality Transformer-GAN for high-quality PET reconstruction49
A 3D multi-modality edge-aware Transformer-GAN for high-quality SPET reconstruction using the corresponding LPET images and T1 acquisitions from magnetic resonance imaging (T1-MRI) is proposed.
- Deep learning based HEp-2 image classification: A comprehensive review47
A comprehensive review of the existing deep learning based HEp-2 cell image classification methods, including core idea, notable achievements, and key strengths and weaknesses of each method are critically analyzed.
- Aucubin Alleviates Seizures Activity in Li-Pilocarpine-Induced Epileptic Mice: Involvement of Inhibition of Neuroinflammation and Regulation of Neurotransmission47
Findings provide the first evidence that AU can exert anti-seizure effects by attenuating gliosis and regulating neurotransmission in epilepsy.
- Mamba-Sea: A Mamba-Based Framework With Global-to-Local Sequence Augmentation for Generalizable Medical Image Segmentation45
This work proposes a novel Mamba-based framework, Mamba-Sea, incorporating global-to-local sequence augmentation to improve the model’s generalizability under domain shift issues, and is the first work to explore the generalization of Mamba for medical image segmentation.
- 43
- Machine learning in medical imaging41
A robust and efficient cascaded shape regression method which iteratively learns the final shape from a given initial shape which performs better than state-of-the-art multi-atlas label fusion based segmentation methods.
- 39
- Interactive medical image segmentation via a point-based interaction39
A novel method for interactive medical image segmentation that has been systematically evaluated in three different segmentation tasks, including CT kidney tumor, MR prostate, and PROMISE12 challenge, showing promising results compared with state-of-the-art methods.
- PoinTramba: A Hybrid Transformer-Mamba Framework for Point Cloud Analysis38
PoinTramba is introduced, a pioneering hybrid framework that synergies the analytical power of Transformer with the remarkable computational efficiency of Mamba for enhanced point cloud analysis and introduces a bi-directional importance-aware ordering strategy to tackle the challenges of random ordering effects.
- Noise-Aware Image Captioning with Progressively Exploring Mismatched Words38
A Noise-aware Image Captioning method to adaptively mitigate the erroneous guidance from noise by progressively exploring mismatched words and can effectively exploit both clean and noisy image-text pairs to learn a more robust mapping function.
- 38
- Single-View 3D Mesh Reconstruction for Seen and Unseen Categories37
This paper proposes an end-to-end two-stage network, GenMesh, to break the category boundaries in reconstruction and introduces a multi-view silhouette loss to supervise the surface generation process, which provides additional regularization and further relieves the overfitting problem.
- Lightweight and effective detection scheme for node clone attack in wireless sensor networks37
A novel scheme to detect the node clone attack in WSN by channel identification characteristic is presented, in which the clone nodes are distinguished by the channel responses between nodes.
- An Efficient Semi-Supervised Framework with Multi-Task and Curriculum Learning for Medical Image Segmentation36
A novel semi-supervised segmentation method via multi-task curriculum learning that yields a much better segmentation performance on a small labeled dataset is presented.
- Constructing hierarchical attentive functional brain networks for early AD diagnosis35
It is argued that the nodes defined by the existing atlases are not necessarily the optimal starting level to build FBN hierarchy and exploring finer nodes may further enrich the FBN representation.
- 34
- 33
- Neural Video Compression with Spatio-Temporal Cross-Covariance Transformers33
This work proposes a new 3D-based transformer module: Spatio-Temporal Cross-Covariance Transformer (ST-XCT), which is integrated into various key coding components of NVC, such as feature extraction, frame reconstruction, and entropy modeling, demonstrating its generalizability.
- 33
- Locality Adaptive Multi-modality GANs for High-Quality PET Image Synthesis33
A locality adaptive multi-modality generative adversarial networks model (LA-GANs) is proposed to synthesize the full-dose PET image from both the low-dose one and the accompanying T1-weighted MRI to incorporate anatomical information for better PET image synthesis.
- 33
- Progressive Cross-Stream Cooperation in Spatial and Temporal Domain for Action Localization32
This work proposes a new progressive cross-stream cooperation (PCSC) framework that improves all three levels of tasks of spatio-temporal action localization, and proposes a new message passing approach to pass information from one stream to another stream, which also leads to better action detection and segment detection models.
- 32
- 31
- Bridging Synthetic and Real Images: A Transferable and Multiple Consistency Aided Fundus Image Enhancement Framework31
This work proposes an end-to-end optimized teacher-student framework to simultaneously conduct image enhancement and domain adaptation, and proposes a novel multi-stage multi-attention guided enhancement network (MAGE-Net) as the backbones of the teacher and student network.
- Learning Discriminative Bayesian Networks from High-Dimensional Continuous Neuroimaging Data31
The contributions of this paper include a new Directed Acyclic Graph (DAG) constraint with theoretical guarantee to ensure the graph validity of GBN, and two general discriminative learning frameworks for Gaussian Bayesian networks (GBN).
- Neural Vector Fields: Implicit Representation by Explicit Learning30
This work proposes a novel 3D representation method, Neural Vector Fields, which not only adopts the explicit learning process to manipulate meshes directly, but also leverages the implicit representation of unsigned distance functions (UDFs) to break the barriers in resolution and topology.
- DC-SSL: Addressing Mismatched Class Distribution in Semi-supervised Learning30
A new SSL learning framework, named Distribution Consistency SSL (DC-SSL), is put forward, which rectifies the pseudolabels from a distribution perspective and revisits the Exponentially Moving Average model and utilizes it to estimate RCD in an iteratively improved manner.
- 30
- 28
- 28
- 27
- 27
- 27
- 27
- 27
- A Fast Algorithm for Creating a Compact and Discriminative Visual Codebook27
Experimental study shows that the compact visual codebook created in this way can achieve excellent classification performance even after a considerable reduction in size.
- 26
- 25
- Group-aware Parameter-efficient Updating for Content-Adaptive Neural Video Compression24
This work introduces a parameter-efficient delta-tuning strategy, which is achieved by integrating several light-weight adapters into each encoding component by using both serial and parallel configuration, and introduces a group-aware approach for updating encoder parameters.
- Integrated Microresonator-Based Microwave Photonic Sensors Assisted by Machine Learning24
The latest developments in the merging of ML and deep learning (DL) with integrated MWP sensors with enhanced sensing capabilities, including athermal operation, resistance to noise and interference, multiple parameters detection, and extended sensing range, are presented.
- CRIF1 as a potential target to improve the radiosensitivity of osteosarcoma24
It is reported that CR6-interacting factor-1 (CRIF1) is highly expressed in osteosarcoma and undergoes nuclear-cytoplasmic shuttling of cyclin-dependent kinase 2 (CDK2) after IR, which may provide insights into the mechanism for other types of radioresistant cancers and be exploited for therapeutic ends.
- 24
- Exploring multifractal‐based features for mild Alzheimer's disease classification24
This work aims to address two issues: (I) if and what multifractal features are sufficiently discriminative to detect AD from the healthy; and (II) if AD classification could be further improved by combining multifractional features with traditional features in this field.
- Machine Learning Techniques for AD/MCI Diagnosis and Prognosis24
This chapter presents some of the latest developments in the application of machine learning techniques to AD and MCI diagnosis and prognosis, and discusses how various biomarkers as well as connectivity networks can be extracted from the various modalities, such as structural T1-weighted imaging, diffusion-tensor imaging (DTI) and resting-state functional magnetic resonance imaging (fMRI).
- Towards Semi-supervised Learning with Non-random Missing Labels23
This work explores the class-level guidance information obtained by the Markov random walk, which is modeled on a dynamically created graph built over the class tracking matrix, and unifies the historical information of class distribution and class transitions caused by the pseudo-rectifying procedure to maintain the model’s unbiased enthusiasm towards assigning pseudo-labels to all classes.
- 23
- Dilated conditional GAN for bone suppression in chest radiographs with enforced semantic features23
An algorithm to suppress the imaging of bone structures within clinical X-ray images, leaving a residual portrayal of lung tissue; such that these images can be used to better serve applications, such as lung nodule detection or pathology based on the radiological reading of chest X-rays.
- Learning Discriminative Stein Kernel for SPD Matrices and Its Applications23
A discriminative SK (DSK), in which an extra parameter vector is defined to adjust the eigenvalues of input SPD matrices, which makes it produce higher classification performance than the original SK and other commonly used methods.
- 22
- 22
- OPML: A one-pass closed-form solution for online metric learning22
OPML and COPML can obtain the promising performance with a very low computational cost and the effectiveness of COPML under the cold start setting is experimentally verified.
- Beyond Covariance: SICE and Kernel Based Visual Feature Representation21
To better deal with the issues of small number of feature vectors and high feature dimensionality, it is proposed to exploit the structure sparsity of visual features and exemplify sparse inverse covariance estimate as a new feature representation and to effectively model complicated feature relationship, to directly compute kernel matrix over feature dimensions.
- Serum NCAM levels and cognitive deficits in first episode schizophrenia patients versus health controls21
The results support the neurodevelopmental dysfunction hypothesis of schizophrenia and suggest that an altered NCAM may be one of the risk factors for schizophrenia including cognitive deficits.
- Hippocampal shape analysis for Alzheimer's disease using an efficient hypothesis test and regularized discriminative deformation21
A framework to perform statistical shape analysis for segmented hippocampi is presented, including an efficient permutation test for detecting subtle class differences, and a regularized discriminative direction method for visualizing the shape discrepancy.
- 20
- 20
- 20
- Efficient radiation dose reduction in whole-brain CT perfusion imaging using a 3D GAN: performance and clinical feasibility19
GAN-based denoising is a promising practical approach for reducing radiation dose in CTP studies and improving lesion characterisation.
- 19
- 18
- 18
- 18
- 17
- Construction of a Collagen-like Protein Based on Elastin-like Polypeptide Fusion and Evaluation of Its Performance in Promoting Wound Healing17
The compound hCol-ELP demonstrated a remarkable capacity to induce the proliferation and migration of mouse embryo fibroblast cells (NIH/3T3), as well as enhance collagen synthesis in human skin fibroblasts (HSF) when tested in vitro.
- 17
- 17
- 17
- Subject-adaptive Integration of Multiple SICE Brain Networks with Different Sparsity17
This paper proposes a subject-adaptive method to integrate multiple SICE networks as a unified representation for classification and provides a unified and effective network representation that is transparent to the sparsity level of Sice networks, and can be readily utilized for further medical analysis.
- Learning Partial Correlation based Deep Visual Representation for Image Classification16
This work obtains a partial correlation based deep visual representation and mitigates the small sample problem often encountered by covariance matrix estimation in CNN.
- 16
- 16
- 16
- A Hybrid LSTM-CPS Approach for Long-Term Prediction of Train Delays in Multivariate Time Series16
A hybrid deep learning solution by integrating long short-term memory (LSTM) and Critical Point Search (CPS) is proposed, demonstrating that new variants outperform the standard LSTM, based on predicting time steps of dwell time feature.
- Network analysis in detection of early-stage mild cognitive impairment16
It is found that the EDD index, better than the widely used CC and EG measures, may serve as an assistant and potential marker for the detection of EMCI.
- Discriminative Brain Effective Connectivity Analysis for Alzheimer's Disease: A Kernel Learning Approach upon Sparse Gaussian Bayesian Network16
This paper employs Fisher kernel to bridge the generative models of sparse Bayesian networks (SBN) and the discriminative classifiers of SVMs, and converts the SBN parameter learning to Fisher kernel learning via minimizing a generalization error bound of SVM.
- 16
- 15
- Improving Weakly Supervised Temporal Action Localization by Exploiting Multi-Resolution Information in Temporal Domain15
This work proposes a two-stage approach to generate high-quality frame-level pseudo labels by fully exploiting multi-resolution information in the temporal domain and complementary information between the appearance and motion streams, and uses a Progressive Temporal Label Refinement framework to iteratively refine the pseudo labels.
- Support Vector Machines for Neuroimage Analysis: Interpretation from Discrimination15
This chapter introduces SVM-based methods and applications for neuroimage analysis from this point of view to characterize the group difference captured by SVM with anatomically interpretable patterns, which provides insights into the unknown mechanism of the brain.
- Enhancing Sample Utilization through Sample Adaptive Augmentation in Semi-Supervised Learning14
Sample adaptive augmentation (SAA) is proposed for this stated purpose and consists of two modules: 1) sample selection module; 2) sample augmentation module, which picks out naive samples based on historical training information at each epoch, and will be augmented in a more diverse manner in the sample augmentation module.
- An efficient radius-incorporated MKL algorithm for Alzheimer׳s disease prediction14
An improved radius-incorporated MKL algorithm for AD prediction is proposed that can better utilise the radius information and achieve higher prediction accuracy than the comparable MKL methods in the literature and demonstrates the highest computational efficiency among all the comparable methods.
- Diffusion Models Need Visual Priors for Image Generation13
This work proposes Diffusion on Diffusion (DoD), an innovative multi-stage generation framework that first extracts visual priors from previously generated samples, then provides rich guidance for the diffusion model leveraging visual priors from the early stages of diffusion sampling.
- Few-shot Unsupervised Domain Adaptation with Image-to-Class Sparse Similarity Encoding13
A novel IMage-to-class sparse Similarity Encoding (IMSE) method that learns SPs to extract the local discriminative information for classification and meanwhile aligns the covariance matrix of the SPs for domain adaptation.
- 13
- Efficiently scaled-up production of recombinant human elastin-like polypeptides using multiple optimization strategies12
This study provides a universal paradigm for the industrial production of ELPs and other similar recombinant proteins, significantly advancing the commercialization of promising ELPs.
- 12
- 12
- 12
- Understanding Masked Autoencoders From a Local Contrastive Perspective12
A local perspective is proposed to explicitly extract a local contrastive form from MAE's reconstructive objective at the patch level and a new empirical framework is introduced, called Local Contrastive MAE (LC-MAE), to analyze both reconstructive and contrastive aspects of MAE.
- 12
- Medxchat: A Unified Multimodal Large Language Model Framework Towards CXRS Understanding and Generation11
This study explores MLLMs for Chest X-Rays through MedXChat, a unified framework enabling seamless interaction for tasks such as text report generation, visual question-answering (VQA), and Text-to-CXR generation.
- UniBrain: A Unified Model for Cross-Subject Brain Decoding11
The UniBrain is presented, a unified brain decoding model that requires no subject-specific parameters and is validated on the brain decoding benchmark, achieving comparable performance to current state-of-the-art subject-specific models with extremely fewer parameters.
- 11
- The NPC1L1 Gene Exerts a Notable Impact on the Reduction of Low-Density Lipoprotein Cholesterol in Response to Hyzetimibe: A Factorial-Designed Clinical Trial11
The g1679C > G SNP in the NPC1L1 gene is critical and exerts a differential impact on the response to hyzetimibe treatment, highlighting the potential necessity for identifying the appropriate subjects to be treated with ezetIMibe.
- Learning Sample-Adaptive Intensity Lookup Table for Brain Tumor Segmentation11
This work proposes to learn a sample-adaptive intensity lookup table (LuT) that adjusts each image’s contrast dynamically so that the resulting images could better serve the subsequent segmentation task.
- 11
- Data Fusion for MaaS: Opportunities and Challenges11
An overview of the recent advances in data fusion for Mobility as a Service (MaaS), including the basics of data fusion theory and the related machine learning methods is provided.
- 11
- 10
- 10
- S-RRG-Bench: Structured radiology report generation with fine-grained evaluation framework10
This work proposes a specialized evaluation metric (S-Score) that not only measures disease prediction accuracy but also evaluates the precision of disease-specific details, thus offering a clinically meaningful metric for report quality that focuses on elements critical to clinical decision-making and demonstrates a stronger alignment with human assessments.
- Rebalanced Vision-Language Retrieval Considering Structure-Aware Distillation10
This paper designs a novel multi-granularity cross-modal matching that incorporates structure-aware distillation alongside the cross-modal matching loss, and proposes a simple yet effective method to rebalance cross-modal matching by learning structure-preserved matching representations.
- MRScore: Evaluating Radiology Report Generation with LLM-based Reward System10
This paper introduces MRScore, an automatic evaluation metric tailored for radiology report generation by leveraging Large Language Models (LLMs), and demonstrates MRScore's higher correlation with human judgments and superior performance in model selection compared to traditional metrics.
- Imbalanced Medical Image Segmentation With Pixel-Dependent Noisy Labels10
The proposed framework, Collaborative Learning with Curriculum Selection (CLCS), addresses pixel-dependent noisy labels with class imbalance by employing a curriculum dynamic thresholding approach adapting to model learning progress to select clean data samples to mitigate the class imbalance issue.
- A deep neural network for positioning and inter-crystal scatter identification in multiplexed PET detectors: a simulation study10
This study indicates that DNN-based event positioning has the potential to enhance 2D coincidence event positioning accuracy by nearly a factor of 3 compared to Anger logic, and further improvements are difficult to foresee without additional information such as event timing.
- Cardiorespiratory dose comparison among six radiotherapy regimens for patients with left-sided breast cancer10
This is the first Bayesian study of several radiotherapy regimens for Breast cancer patients on the left side, and the findings can be used to select appropriate radiotherapy programs for breast cancer patients.
- Influenza vaccination coverage among health-care workers during the COVID-19 epidemic in 2020/2021 influenza season: Evidence from a web-based survey in northwestern China10
Under the policy of voluntary and self-pay influenza vaccination in Ningxia, the coverage rate of influenza vaccine among health-care workers was far below the vaccination requirements of flu vaccine in influenza season even during the COVID-19 epidemic.
- 10
- Integrating Functional and Structural Connectivities via Diffusion-Convolution-Bilinear Neural Network9
A Diffusion-Convolution-Bilinear Neural Network (DCB-NN) framework for brain network analysis, which couples FC and SC seamlessly and considers wider interactions among network nodes and demonstrates the effectiveness and advantages of this proposed method.
- Multifractal analysis of white matter structural changes on 3D magnetic resonance imaging between normal aging and early Alzheimer’s disease9
The proposed IRBCMA can be an alternative and more accurate algorithm for 3D volume analysis and highlight the potential usefulness of multifractal analysis, which may contribute to clarify some aspects of the etiology of AD through detection of structural changes in white matter.
- 9
- WDM optical network testbed and distributed storage application9
It is demonstrated that incorporating a lightweight data transport protocol with optical network control could be an efficient approach for data-intensive storage applications by exploiting the dynamic connection establishment and backup protection features.
- High-gradient microstructured hybrid microfluidic chip for rare tumor cell capture8
A novel chip structure design for tumor cell capture is presented, which holds promise for improving the capture of tumor cells in complex biological samples and in stabilizing the captured cancer cells, enhancing separation efficiency.
- 8
- Modelling diffusion process by deep neural networks for image retrieval8
This paper proposes to treat diffusion process as a “black box” and directly model it by training deep neural networks, so as to obtain better image representation that assimilates the effect of diffusion process and works with Euclidean search.
- Particle Swarm Optimization Model of Distributed Network Planning8
This study presents the optimal planning algorithm basing on particle swarm, which can ensure the operational efficiency, stability and security of the network and save a lot of hardware resources to reasonably plan the distributed computer network.
- 8
- Diversity-Enhanced Collaborative Mamba for Semi-Supervised Medical Image Segmentation7
A novel Diversity-enhanced Collaborative Mamba framework (namely DCMamba) for semi-supervised medical image segmentation, which explores and utilizes the diversity from data, network, and feature perspectives, and adopts an uncertainty-weighted contrastive learning mechanism to enhance the diversity of feature representation.
- Double promoter and tandem gene strategy for efficiently expressing recombinant FGF217
E engineered strains that exhibit a high yield of FGF21 are successfully engineered, revealing that vectors with various copy numbers significantly impact the expression of FGF21, and the results showed a 4.35-fold increase in expression levels.
- Cross-Modal Attention-Guided Convolutional Network for Multi-modal Cardiac Segmentation7
A novel cross-modal attention-guided convolutional network for multi- modal cardiac segmentation that outperforms the baselines in terms of the segmentation performance and can be trained in an end-to-end manner.
- A Graph-Embedding Approach to Hierarchical Visual Word Mergence7
This paper proposes a unified hierarchical merging approach built upon the graph-embedding framework that can maintain excellent image classification performance even after a significant dimension reduction, and outperform all the existing comparable visual word-merging methods.
- A Study of Hippocampal Shape Difference Between Genders by Efficient Hypothesis Test and Discriminative Deformation7
The Fisher permutation and bootstrap tests are used to differentiate hippocampal shape between genders and a "informative direction" is found to preserve the original discriminative information.
- Stereoscopic visualization and editing of automatic abdominal aortic aneurysms (AAA) measurements for stent graft planning7
A system for automatic AAA measurement and interactive 3D editing that can be reviewed and edited interactively in the 3D context of the volumetric rendering of the aorta, so that relationships of the measurements and theAorta are clearly perceived.
- 6
- Reflective microring-resonator-based microwave photonic sensor incorporating a self-attention assisted convolutional neural network6
A reflective microring resonator (MRR)-based microwave photonic (MWP) sensor incorporating a self-attention convolutional neural network (CNN) is presented, showing a 10-fold improvement in sensing accuracy compared to that of the traditional linear fitting model.
- Task-Oriented Multi-Modal Mutual Leaning for Vision-Language Models6
This work proposes a class-aware text prompt (CTP), which can effectively involve image semantics and avoid introducing extra ambiguities into different prompts, and proposes text-guided feature tuning (TFT) to make the image branch attend to class-related representation.
- 6
- 6
- The expression of G protein-coupled receptor kinase 5 and its interaction with dendritic marker microtubule-associated protein-2 after status epilepticus6
GRK5 was involved in the development of epilepsy and was associated with dendritic formation in epilepsy and the concrete mechanisms of the GRK5 within epileptogenesis require further research.
- 6
- 5
- 5
- Entropy-based Optimization on Individual and Global Predictions for Semi-Supervised Learning5
This paper proposes a new SSL method that employs Entropy-based optimization on both Individual and Global predictions of unlabeled instances, dubbed EntInG and shows that current dominant SSL methods can be viewed as an implicit form of IPEM improved by recent augmentation techniques.
- 5
- Multiple Obstacle Detection for Assistance Driver System Using Deep Neural Networks5
A deep convolutional networks framework for solving the online multiple obstacles detection problem and is trained in an end-to-end manner for better adapting the influences of visual information as well as inter-object relations and orientation information.
- 5
- 5
- Hippocampal Shape Classification Using Redundancy Constrained Feature Selection5
This work identifies the drawbacks of SVM-RFE, and proposes a novel class-separability-based feature selection approach to overcome them, and formulate feature selection as a constrained integer optimization and develop a new algorithm to efficiently and optimally solve this problem.
- 5
- 5
- 4
- 4
- DiN: Diffusion Model for Robust Medical VQA with Semantic Noisy Labels4
The first benchmark for noisy labels in Med-VQA is established by simulating human mislabeling with semantically designed noise types, and the DiN framework is introduced, which leverages a diffusion model to handle noisy labels in Med-VQA.
- InfGen: A Resolution-Agnostic Paradigm for Scalable Image Synthesis4
This work presents the InfGen, replacing the VAE decoder with the new generator, for generating images at any resolution from a fixed-size latent without retraining the diffusion models, which simplifies the process, reducing computational complexity and can be applied to any model using the same latent space.
- Neural Vector Fields: Generalizing Distance Vector Fields by Codebooks and Zero-Curl Regularization4
A novel 3D representation, Neural Vector Fields (NVF), which adopts the explicit learning process to manipulate meshes and implicit unsigned distance function (UDF) representation to break the barriers in resolution and topology and incorporates two types of shape codebooks, i.e., NVFs (Lite or Ultra), to promote cross-category reconstruction through encoding cross-object priors.
- Wide-Range Operation of Microwave Photonic Sensor Using Recurrent Neural Network4
A microwave photonic sensor whose operational range and sensing accuracy are enhanced through the utilization of a recurrent neural network (RNN) is presented whose operational range and sensing accuracy are enhanced through the utilization of a recurrent neural network.
- Progressive Classifier and Feature Extractor Adaptation for Unsupervised Domain Adaptation on Point Clouds4
This work proposes a novel framework that deeply couples the classifier and feature extractor adaption for 3D UDA, dubbed Progressive Classifier and Feature Extractor Adaptation (PCFEA).
- Towards Speech Classification from Acoustic and Vocal Tract data in Real-time MRI4
A multimodal phonemic classifier that combines articulatory data with speech audio features to improve performance and analyze the model and discuss the phonetic implications.
- Vision-Language Models for Biomedical Applications4
An overview of the inaugural edition of the VLM4Bio workshop is provided, summarizing the key discussions, contributions, and future directions for expanding the workshop's scope and influence in subsequent editions.
- Neuroimaging signatures and a deep learning modeling for early diagnosing and predicting non-pharmacological therapy success for subclinical depression comorbid sleep disorders in college students4
The HFBN model demonstrated superior performance over traditional deep learning models, accurately predicting therapeutic outcomes and diagnosing subclinical depression, achieving cumulative scores of 80.47% for sleep quality prediction and 84.67% for depression prediction, along with an overall diagnostic accuracy of 82.34%.
- 4
- 4
- 4
- Does perfect filtering really guarantee perfect phase correction for diffusion MRI data?4
It is argued that even a perfect image filter is insufficient to produce perfect phase correction, and a complex polar coordinate system (CPCS) is proposed to analyze its procedures in detail and find that phase correction has not sufficiently utilized the background phase, and thus proposes a quantitative criterion to fully exploit the Background phase.
- 4
- Special Issue on High Performance Computing in Bio-medical Informatics4
This special issue presents the state of the art of computational methods that address the above challenges of high performance computing in bio-medical informatics, and proposes deep neural networks and graphical models to address a variety of neuroscience related problems investigated using multiple imaging modalities.
- Vulnerability of calbindin-positive interneurons to status epilepticus varies in different regions of rat hippocampus4
It was shown that the sensitivity of CB-positive interneurons to SE varied in different regions of hippocampus, including hippocampal CA1, CA3, dentate gyrus and hilus.
- MR-Less Surface-Based Amyloid Estimation by Subject-Specific Atlas Selection and Bayesian Fusion4
Qualitative and quantitative comparison with and without MRI are presented and it is proposed that using PET only, the average error on the brain surface was around 13% compared to MRI-dependant method.
- Regularized Discriminative Direction for Shape Difference Analysis4
A regularized discrim inative direction is developed by requiring a shape to conform to its population distribution when it deforms along the discriminative direction, and shows superior performance in detecting and localizing the difference of hippocampal shapes for sex.
- AWF: Adaptive Weight Fusion for Enhanced Class Incremental Semantic Segmentation3
Adaptive Weight Fusion (AWF), which introduces an alternating training strategy for the fusion parameter, allowing for more flexible and adaptive weight integration, significantly improving results on benchmark CISS tasks compared to the original EWF.
- 3
- 3
- Clean Label Disentangling for Medical Image Segmentation with Noisy Labels3
This work comes up with a simple but efficient class-balanced sampling strategy to tackle the class-imbalanced problem, which enables the newly proposed clean label disentangling framework to successfully select clean labels from the given label sets and encourages the model to learn from the correct annotations.
- Deep Learning Models with Applications to Brain Image Analysis3
This chapter introduces the pre-processing steps for brain images and the fundamental concepts of deep neural networks, as well as their applications in brain image classification, segmentation, registration, and image synthesis/augmentation.
- 3
- 3
- 3
- 3
- Application of SVMs to the Bag-of-Features Model: A Kernel Perspective3
The fundamentals of the Bag-of-features model in image categorisation are introduced and the novel kernels developed to compare images based on a variety of representations incurred by this model are shown.
- Max-Margin Based Learning for Discriminative Bayesian Network from Neuroimaging Data3
A learning approach to constructing SGBNs that are both representative and discriminative for groups in comparison is proposed, and a max-margin criterion built directly upon the SGBN models is proposed to effectively optimize the classification performance of the S GBNs.
- 2
- 2
- 2
- RadReason: Radiology Report Evaluation Metric with Reasons and Sub-Scores2
RadReason is introduced, a novel evaluation framework for radiology reports that not only outputs fine-grained sub-scores across six clinically defined error types, but also produces human-readable justifications that explain the rationale behind each score.
- A renaissance of explicit motion information mining from transformers for action recognition2
In EMIM, the desirable affinity matrix is constructed in a cost volume style, where the set of key candidate tokens is sampled from the query-based neighboring area in the next frame in a sliding-window manner and is used to aggregate contextual information for appearance modeling and is converted into motion features for motion modeling as well.
- 2
- A novel x-Ray and γ-Ray combination strategy for radiotherapy after breast-conserving surgery in patients with right breast cancer2
Overall, for the TaiChiB dual-modality radiotherapy system, the radiation dose outside the target region decreases rapidly, thereby minimizing radiation exposure to neighboring organs and ensuring the conformity of the target area.
- 2
- Exploring the biological basis of CT imaging features in pancreatic neuroendocrine tumors: a two-center study2
The 3D imaging features showed the strongest relationships with clinical factors and immunohistochemical markers, and the 3D feature-based signatures showed optimal performance in clinical factor prediction.
- Rethinking Data Perturbation and Model Stabilization for Semi-supervised Medical Image Segmentation2
DPMS is proposed, which adopts a plain teacher-student framework with a standard supervised loss and unsupervised consistency loss and can obtain new state-of-the-art performance on the public 2D ACDC and 3D LA datasets across various semi-supervised settings.
- 2
- 2
- A Novel Image-Specific Transfer Approach for Prostate Segmentation in MR Images2
A novel transfer approach for prostate segmentation in MR images using a pair of dictionaries and a mapping matrix is proposed by a novel Semi-Coupled Dictionary Transfer Learning (SCDTL).
- 2
- 2
- 1
- 1
- Automated Medical Report Generation and Visual Question Answering1
This presentation will showcase the latest research on automated medical diagnostic report generation and medical visual question answering, highlighting how the work has transitioned from traditional encoder-decoder models to cutting-edge approaches utilizing large language models (LLMs).
- 1
- 1
- Medical Imaging Based Diagnosis Through Machine Learning and Data Analysis1
This chapter introduces several of the recent achievements with machine learning methods for feature extraction and representation, classification, dense prediction, segmentation and synthesis, and multi-modality analysis, across the pipeline of computer-assisted diagnosis (CAD).
- 1
- 1
- 1
- 1
- Exploring Compact Representation of SICE Matrices for Functional Brain Network Classification1
To cater for the requirement of both discrimination and interpretation in neuroimage analysis, a novel pre-image estimation algorithm is developed to make the obtained connectivity components anatomically interpretable.
- 1
- Modification of improved decoding algorithm for LDPC codes over BEC1
A modification of the improved decoding algorithm for low-density parity-check (LDPC) codes over the binary erasure channel (BEC) is presented and it decreases error rate considerably, whereas having almost the same complexity as the message passing algorithm.
- Research and development of Multi- Agent System based agile collaborative design system for machining centre1
An assistant-software has a certain degree of intelligent ability will have more advantages to improve the design quality and short design cycle simultaneously.
- Reconstructing Surface Discontinuities by Intersecting Tangent Planes of Advancing Mesh Frontiers1
This work has shown that the polygonization and surface fitting approaches to surface reconstruction face the same chicken-and-egg problem, so the algorithm needs to know the locations and orientations of edge discontinuities, which are not directly available in the unorganized point set.
- ACQUIRING 3D MODELS FROM IMAGES FOR MULTIMEDIA SYSTEMS1
This paper presents a system for automated acquisition of 3D models and presents a new algorithm for constructive polygonization of3D points and highlights the main components of the system.
- –
- –
- Medical Referring Image Segmentation via Next-Token Mask Prediction–
NTP-MRISeg is proposed, a novel framework that reformulates MRIS as an autoregressive next-token prediction task over a unified multimodal sequence of tokenized image, text, and mask representations, and enables the use of pretrained tokenizers from emerging large-scale multimodal models, enhancing generalization and adaptability.
- Naturalistic Social Dyads Assessment In Free Play: A Multi-view Framework For Recognizing Co-located Eye Contact Between Children and Their Assessors–
It is suggested that multi-view video analysis can objectively identify clinically defined eye contact behaviors among multiple individuals during clinical free play assessments, and may support scalable behavioral coding and quantitative assessment of clinically relevant social behaviors consistent with established criteria.
- –
- –
- A Swin-Transformer enhanced LLM framework for medical visual question answering–
This work proposes a novel LLM-based framework using the Llama3.1-8B model as the backbone and demonstrates the effectiveness of the approach in medical image recognition, natural language understanding, and accurate answer generation.
- –
- Constructing a Unified Vision-Language Model for Chest Radiograph–based Diagnostics, Medical Education, and Data Augmentation–
MedXChat integrated report generation, VQA, and image synthesis within a unified framework, achieving state-of-the-art performance and may support future professional applications and enhance radiologic workflows, education, and data augmentation.
- –
- Unifying Generative Self-Supervised Paradigms with Diffusion Models–
This work proposes a novel Generative framework with a UNified Self-supervised training paradigm (GUNS) to learn semantic information at different granularities and unify multiple image generation tasks within a denoising diffusion model.
- –
- –
- Microwave photonics assisted by machine learning–
Recent advances are presented with a focus on using machine learning and deep learning techniques to assist microwave photonics in sensing with thermal interference mitigation and multi-parameter sensing capabilities.
- –
- –
- –
- Instance Image Retrieval by Aggregating Sample-based Discriminative Characteristics–
This paper proposes a new query expansion method that needs neither pseudo-labeled negatives nor the online learning process, leading to more efficient retrieval and even better performance, and demonstrates its application to the state-of-the-art diffusion-based image retrieval.
- –
- –
- –
- O4‐07‐04: AUTOMATED REPORTING OF AMYLOID PET QUANTIFICATION ON BRAIN SURFACE THROUGH A WEB INTERFACE–
The results suggest that selecting subjects having AD pathology based on clinical criteria remains a challenge, enrichment by using amyloid PET imaging is effective and feasible, and enrichment by assessingAmyloid plaque burden is critically important to clinical studies in early stages of AD because of a higher incidence of negative amyloids findings.
- Routing Strategy Based on Local Density Sensing in Delay Tolerant Network–
This paper has been found to be in violation of Academy Publisher’s Publication Principles and reasonable effort should be made to remove all past references to this paper.
- –
- –
- –
- –
- A Static Measurement Model for Trustworthiness of Software’s Behaviors–
A static measuring model for trustworthiness of software behaviors was presented, of which the strong constraint condition of the static trustworthinessof software behaviors in the model is brought forward firstly as the idealized direction and then the weak condition was put forward for engineering-oriented application.
- –
- –
- Discriminative Maximum Margin Image Object Categorization with Exact Inference–
This work explicitly model object dependencies in a sparse graphical topology induced by the adjacency of objects in the image, which benefits inference, and uses maximum margin principle to learn the model discriminatively.
- –
- Priority-based error correction using turbo codes for compressed AIRS data–
This paper formulates a priority scheme for the compressed data and presents an error correction approach based on minimizing impact on the retrieval products, and demonstrates that this approach can achieve negligible product degradation while maintaining an overall 3-to-1 compression ratio on the final file.
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-14. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.