2026
Multimodal Feature Prototype Learning for Interpretable and Discriminative Cancer Survival Prediction
IEEE Journal of Biomedical and Health Informatics 2026 JCR Q1
This work improves cancer survival prediction by linking histopathological patterns with genomic information in an interpretable prototype-learning framework. Beyond predictive accuracy, it provides transparent patient-level evidence of the tumour phenotypes associated with survival outcomes, supporting more trustworthy prognosis and personalised decision-making.
Multimodal Feature Prototype Learning for Interpretable and Discriminative Cancer Survival Prediction
IEEE Journal of Biomedical and Health Informatics 2026 JCR Q1
This work improves cancer survival prediction by linking histopathological patterns with genomic information in an interpretable prototype-learning framework. Beyond predictive accuracy, it provides transparent patient-level evidence of the tumour phenotypes associated with survival outcomes, supporting more trustworthy prognosis and personalised decision-making.
3DGFA: 3D Gaze-consistent Face Anonymization
ACM International Conference on Multimedia (ACM MM) 2026 CORE A* CCF-A
This work advances face anonymization by preserving coherent 3D gaze behaviour across different viewpoints, ensuring that privacy protection does not compromise visual cue for attention and interaction. It enables anonymized face images to remain useful for gaze-sensitive applications while maintaining strong identity protection, facial attributes, robustness, and generalization.
3DGFA: 3D Gaze-consistent Face Anonymization
ACM International Conference on Multimedia (ACM MM) 2026 CORE A* CCF-A
This work advances face anonymization by preserving coherent 3D gaze behaviour across different viewpoints, ensuring that privacy protection does not compromise visual cue for attention and interaction. It enables anonymized face images to remain useful for gaze-sensitive applications while maintaining strong identity protection, facial attributes, robustness, and generalization.
3D-Aware Eye Generation for Controllable Gaze Data Synthesis
ACM International Conference on Multimedia (ACM MM) 2026 CORE A* CCF-A
This work develops a 3D-aware generative framework for controllable gaze synthesis. It models gaze through explicit eyeball rotation in canonical 3D space, enabling diverse image generation with physically consistent eye motion and reliable gaze labels. Its few-shot adaptation strategy further transfers target-domain appearance from a small number of unlabeled images, producing synthetic data that consistently improve downstream gaze estimation.
3D-Aware Eye Generation for Controllable Gaze Data Synthesis
ACM International Conference on Multimedia (ACM MM) 2026 CORE A* CCF-A
This work develops a 3D-aware generative framework for controllable gaze synthesis. It models gaze through explicit eyeball rotation in canonical 3D space, enabling diverse image generation with physically consistent eye motion and reliable gaze labels. Its few-shot adaptation strategy further transfers target-domain appearance from a small number of unlabeled images, producing synthetic data that consistently improve downstream gaze estimation.
MedGSSR: Generalizable Medical Image Super-Resolution 3D Reconstruction via Hierarchical Feed-forward Gaussian Splatting
European Conference on Computer Vision (ECCV) 2026 CORE A* CCF-B
This work develops MedGSSR, an end-to-end feed-forward framework for arbitrary-scale 3D medical image super-resolution. By representing MRI and CT volumes with explicit 3D Gaussian fields, it enables fast, anatomically faithful, and high-detail reconstruction while generalising to unseen datasets without per-subject optimisation.
MedGSSR: Generalizable Medical Image Super-Resolution 3D Reconstruction via Hierarchical Feed-forward Gaussian Splatting
European Conference on Computer Vision (ECCV) 2026 CORE A* CCF-B
This work develops MedGSSR, an end-to-end feed-forward framework for arbitrary-scale 3D medical image super-resolution. By representing MRI and CT volumes with explicit 3D Gaussian fields, it enables fast, anatomically faithful, and high-detail reconstruction while generalising to unseen datasets without per-subject optimisation.
Distance-Aware Joint Spatio-Temporal Graph Contrastive Learning for Major Depressive Disorder Diagnosis
Submitted to IEEE Journal of Biomedical and Health Informatics (JBHI). Under review. JCR Q1
This work develops a spatio-temporal graph contrastive learning framework that models dynamic functional brain connectivity from resting-state fMRI for Major Depressive Disorder diagnosis. It aims to improve imaging-assisted diagnosis by learning reliable and interpretable spatio-temporal brain network representations.
Distance-Aware Joint Spatio-Temporal Graph Contrastive Learning for Major Depressive Disorder Diagnosis
Submitted to IEEE Journal of Biomedical and Health Informatics (JBHI). Under review. JCR Q1
This work develops a spatio-temporal graph contrastive learning framework that models dynamic functional brain connectivity from resting-state fMRI for Major Depressive Disorder diagnosis. It aims to improve imaging-assisted diagnosis by learning reliable and interpretable spatio-temporal brain network representations.
fMRI-Diffusion: Generating fMRI Time Series Via a Temporal Transformer Diffusion Model for Major Depressive Disorder Diagnosis
Submitted to Computer Methods and Programs in Biomedicine. Under review. JCR Q1
This work proposes a generative framework for synthesizing resting-state fMRI time series to improve Major Depressive Disorder diagnosis. It provides a practical data augmentation strategy for limited neuroimaging datasets while preserving clinically meaningful temporal brain activity patterns.
fMRI-Diffusion: Generating fMRI Time Series Via a Temporal Transformer Diffusion Model for Major Depressive Disorder Diagnosis
Submitted to Computer Methods and Programs in Biomedicine. Under review. JCR Q1
This work proposes a generative framework for synthesizing resting-state fMRI time series to improve Major Depressive Disorder diagnosis. It provides a practical data augmentation strategy for limited neuroimaging datasets while preserving clinically meaningful temporal brain activity patterns.
Neuroscience-inspired Staged Representation Learning with Disentangled Coarse-and Fine-Grained Semantics for EEG Visual Decoding
Submitted to Neurocomputing. Under review. JCR Q1
This work improves EEG-based visual decoding by modelling how the human visual system processes information in stages. This enables more accurate and robust zero-shot visual decoding, supporting brain-computer interfaces, neural signal interpretation, and AI-assisted neurorehabilitation.
Neuroscience-inspired Staged Representation Learning with Disentangled Coarse-and Fine-Grained Semantics for EEG Visual Decoding
Submitted to Neurocomputing. Under review. JCR Q1
This work improves EEG-based visual decoding by modelling how the human visual system processes information in stages. This enables more accurate and robust zero-shot visual decoding, supporting brain-computer interfaces, neural signal interpretation, and AI-assisted neurorehabilitation.
MDeiT: A lightweight and explainable hybrid model for cancer classification in histopathology images
Biomedical Signal Processing and Control 2026 JCR Q1
This work develops a lightweight and interpretable AI model for cancer classification in histopathology images. It enables accurate, efficient, and explainable cancer classification. Its visual explanations are validated by pathologist, supporting more transparent and clinically reliable AI-assisted histopathology analysis.
MDeiT: A lightweight and explainable hybrid model for cancer classification in histopathology images
Biomedical Signal Processing and Control 2026 JCR Q1
This work develops a lightweight and interpretable AI model for cancer classification in histopathology images. It enables accurate, efficient, and explainable cancer classification. Its visual explanations are validated by pathologist, supporting more transparent and clinically reliable AI-assisted histopathology analysis.
SR3R: Rethinking Super-Resolution 3D Reconstruction With Feed-Forward Gaussian Splatting
The 43th IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2026 CORE A* CCF-A
This work enables feed-forward and generalisable 3D super-resolution reconstruction from low-resolution multi-view images, avoiding costly per-scene optimisation. By learning a direct mapping to high-resolution 3D Gaussian representations, SR3R can reconstruct unseen scenes efficiently while improving visual detail, geometric consistency, and cross-scene generalisation.
SR3R: Rethinking Super-Resolution 3D Reconstruction With Feed-Forward Gaussian Splatting
The 43th IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2026 CORE A* CCF-A
This work enables feed-forward and generalisable 3D super-resolution reconstruction from low-resolution multi-view images, avoiding costly per-scene optimisation. By learning a direct mapping to high-resolution 3D Gaussian representations, SR3R can reconstruct unseen scenes efficiently while improving visual detail, geometric consistency, and cross-scene generalisation.
NCSTR: Node-Centric Decoupled Spatio-Temporal Reasoning for Video-based Human Pose Estimation
The 43th IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2026 CORE A* CCF-A
This work improves video-based human pose estimation under motion blur, occlusion, and complex movement. The method achieves state-of-the-art performance and supports applications in human motion analysis, sports analytics, and intelligent visual systems.
NCSTR: Node-Centric Decoupled Spatio-Temporal Reasoning for Video-based Human Pose Estimation
The 43th IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2026 CORE A* CCF-A
This work improves video-based human pose estimation under motion blur, occlusion, and complex movement. The method achieves state-of-the-art performance and supports applications in human motion analysis, sports analytics, and intelligent visual systems.
Prompt-Free Lightweight SAM Adaptation for Histopathology Nuclei Segmentation with Strong Cross-Dataset Generalization
IEEE 23rd International Symposium on Biomedical Imaging (ISBI) 2026 CORE/ERA2010 A
This work improves the practicality of foundation models for computational pathology. It enables accurate and efficient nuclei segmentation across diverse histopathology images, supports scalable tissue analysis for cancer diagnosis, biomarker quantification, and prognosis prediction.
Prompt-Free Lightweight SAM Adaptation for Histopathology Nuclei Segmentation with Strong Cross-Dataset Generalization
IEEE 23rd International Symposium on Biomedical Imaging (ISBI) 2026 CORE/ERA2010 A
This work improves the practicality of foundation models for computational pathology. It enables accurate and efficient nuclei segmentation across diverse histopathology images, supports scalable tissue analysis for cancer diagnosis, biomarker quantification, and prognosis prediction.
IE-SRGS: An Internal-External Knowledge Fusion Framework for High-Fidelity 3D Gaussian Splatting Super-Resolution
The 40th Annual AAAI Conference on Artificial Intelligence (AAAI) 2026 CORE A* CCF-A
This work enables high-fidelity 3D reconstruction from low-resolution inputs, improving both visual quality and geometric consistency. It introduces a 3D Gaussian Splatting super-resolution framework that fuses external 2D super-resolution and depth priors with internal multi-scale 3DGS representations.
IE-SRGS: An Internal-External Knowledge Fusion Framework for High-Fidelity 3D Gaussian Splatting Super-Resolution
The 40th Annual AAAI Conference on Artificial Intelligence (AAAI) 2026 CORE A* CCF-A
This work enables high-fidelity 3D reconstruction from low-resolution inputs, improving both visual quality and geometric consistency. It introduces a 3D Gaussian Splatting super-resolution framework that fuses external 2D super-resolution and depth priors with internal multi-scale 3DGS representations.
MVTT-GMamba: A Multimodal Graph Reasoning Framework for Anti-VEGF Efficacy Prediction in Diabetic Macular Edema
IEEE Journal of Biomedical and Health Informatics 2026 JCR Q1
We propose a multimodal AI solution that combines OCT images and clinical indicators to predict Anti-VEGF treatment response in diabetic macular edema. This work supports more personalised ophthalmic care by helping clinicians anticipate treatment efficacy and make better informed therapeutic decisions.
MVTT-GMamba: A Multimodal Graph Reasoning Framework for Anti-VEGF Efficacy Prediction in Diabetic Macular Edema
IEEE Journal of Biomedical and Health Informatics 2026 JCR Q1
We propose a multimodal AI solution that combines OCT images and clinical indicators to predict Anti-VEGF treatment response in diabetic macular edema. This work supports more personalised ophthalmic care by helping clinicians anticipate treatment efficacy and make better informed therapeutic decisions.
DCG-Net: Dual Cross-Attention with Concept-Value Graph Reasoning for Interpretable Medical Diagnosis
The 34th IEEE International Conference on Multimedia and Expo (ICME) 2026 CORE A CCF-B
This work develops an interpretable medical diagnosis framework that links image evidence with clinical concepts, and models their relationships through graph reasoning. It improves diagnostic performance while producing clinically meaningful explanations, supporting trustworthy AI-assisted medical image analysis.
DCG-Net: Dual Cross-Attention with Concept-Value Graph Reasoning for Interpretable Medical Diagnosis
The 34th IEEE International Conference on Multimedia and Expo (ICME) 2026 CORE A CCF-B
This work develops an interpretable medical diagnosis framework that links image evidence with clinical concepts, and models their relationships through graph reasoning. It improves diagnostic performance while producing clinically meaningful explanations, supporting trustworthy AI-assisted medical image analysis.
Causal-Audit: Explicit and Auditable Graph-based Reasoning via Target-Aware Causal Chain Construction
The 64th Annual Meeting of the Association for Computational Linguistics (ACL) 2026 CORE A* CCF-A
We propose a target-aware framework that performs explicit causal reasoning for intervention-based question answering. By constructing auditable causal chains and aggregating path-level evidence, the method improves the interpretability, robustness, and trustworthiness of LLM reasoning under complex interventions.
Causal-Audit: Explicit and Auditable Graph-based Reasoning via Target-Aware Causal Chain Construction
The 64th Annual Meeting of the Association for Computational Linguistics (ACL) 2026 CORE A* CCF-A
We propose a target-aware framework that performs explicit causal reasoning for intervention-based question answering. By constructing auditable causal chains and aggregating path-level evidence, the method improves the interpretability, robustness, and trustworthiness of LLM reasoning under complex interventions.
Real-Time Thoracic Duct Visualization in Lung Cancer Surgery: Feasibility of Indocyanine Green Fluorescence Imaging via Inguinal Lymph Node Injection
Journal of Thoracic Disease 2026
We explored real-time thoracic duct visualization in lung cancer surgery using indocyanine green fluorescence imaging via inguinal lymph node injection. This technique may improve intraoperative anatomical guidance and help reduce thoracic duct injury and postoperative chyle leakage.
Real-Time Thoracic Duct Visualization in Lung Cancer Surgery: Feasibility of Indocyanine Green Fluorescence Imaging via Inguinal Lymph Node Injection
Journal of Thoracic Disease 2026
We explored real-time thoracic duct visualization in lung cancer surgery using indocyanine green fluorescence imaging via inguinal lymph node injection. This technique may improve intraoperative anatomical guidance and help reduce thoracic duct injury and postoperative chyle leakage.
2025
Kf-Gs: Kalman Filter-Guided Gaussian Splatting for Real-Time High-Quality Dynamic Scene Reconstruction
Journal of Visual Communication and Image Representation 2025
This work introduces a Kalman filter guided Gaussian Splatting method for real-time dynamic scene reconstruction. It improves motion estimation and temporal stability in dynamic novel view synthesis, enabling smoother and more coherent rendering for applications such as VR, AR, and interactive 4D visualization.
Kf-Gs: Kalman Filter-Guided Gaussian Splatting for Real-Time High-Quality Dynamic Scene Reconstruction
Journal of Visual Communication and Image Representation 2025
This work introduces a Kalman filter guided Gaussian Splatting method for real-time dynamic scene reconstruction. It improves motion estimation and temporal stability in dynamic novel view synthesis, enabling smoother and more coherent rendering for applications such as VR, AR, and interactive 4D visualization.
A Component-Guided Frequency Perturbation Approach for Privacy Preserving Face Recognition
International Conference on Machine Learning and Computer Application (ICMLCA) 2025
This work presents a practical privacy-preserving face recognition method that protects identity information through component-guided frequency perturbation. It supports safer deployment of biometric systems by reducing reconstruction risks while maintaining reliable recognition performance.
A Component-Guided Frequency Perturbation Approach for Privacy Preserving Face Recognition
International Conference on Machine Learning and Computer Application (ICMLCA) 2025
This work presents a practical privacy-preserving face recognition method that protects identity information through component-guided frequency perturbation. It supports safer deployment of biometric systems by reducing reconstruction risks while maintaining reliable recognition performance.
ViewCloud: A lightweight multi-view point cloud representation for efficient 3D recognition and cross-domain retrieval
Computer-Aided Design 2025
This work develops ViewCloud as a lightweight 3D representation for practical 3D recognition and cross-domain retrieval. By combining the efficiency of point clouds with the semantic richness of multi-view renderings, it supports low-cost, scalable, and deployment-friendly 3D understanding for CAD, robotics, and real-world visual retrieval applications.
ViewCloud: A lightweight multi-view point cloud representation for efficient 3D recognition and cross-domain retrieval
Computer-Aided Design 2025
This work develops ViewCloud as a lightweight 3D representation for practical 3D recognition and cross-domain retrieval. By combining the efficiency of point clouds with the semantic richness of multi-view renderings, it supports low-cost, scalable, and deployment-friendly 3D understanding for CAD, robotics, and real-world visual retrieval applications.
Explainable artificial intelligence in biomedical image analysis: A comprehensive survey
Submitted to ACM Computing Survey. Under review. 2025 JCR Q1
This work supports the engineering of transparent and clinically trustworthy AI systems by helping researchers select suitable explanation methods, evaluate interpretability, and align model outputs with biomedical decision-making needs.
Explainable artificial intelligence in biomedical image analysis: A comprehensive survey
Submitted to ACM Computing Survey. Under review. 2025 JCR Q1
This work supports the engineering of transparent and clinically trustworthy AI systems by helping researchers select suitable explanation methods, evaluate interpretability, and align model outputs with biomedical decision-making needs.
Deep Learning Model Inversion Attacks and Defenses: A Comprehensive Survey
Artificial Intelligence Review 2025 JCR Q1
This work provides practical insights for engineering secure and privacy-preserving AI systems that can resist sensitive data reconstruction attacks.
Deep Learning Model Inversion Attacks and Defenses: A Comprehensive Survey
Artificial Intelligence Review 2025 JCR Q1
This work provides practical insights for engineering secure and privacy-preserving AI systems that can resist sensitive data reconstruction attacks.
AI-Driven Household Electricity Load Forecasting: Challenges, Methods, and Future Directions
Artificial Intelligence Review 2025 JCR Q1
This work provides practical guidance for building intelligent forecasting systems that support smart grids, battery optimisation, demand response, and renewable energy integration.
AI-Driven Household Electricity Load Forecasting: Challenges, Methods, and Future Directions
Artificial Intelligence Review 2025 JCR Q1
This work provides practical guidance for building intelligent forecasting systems that support smart grids, battery optimisation, demand response, and renewable energy integration.
Explainable multimodal hematology analysis for white blood cell classification and attribute prediction
Computers in Biology and Medicine 2025 JCR Q1
This work develops an explainable vision-language model for automated white blood cell analysis. This supports efficient and transparent AI-assisted blood cell assessment for clinical and biomedical applications.
Explainable multimodal hematology analysis for white blood cell classification and attribute prediction
Computers in Biology and Medicine 2025 JCR Q1
This work develops an explainable vision-language model for automated white blood cell analysis. This supports efficient and transparent AI-assisted blood cell assessment for clinical and biomedical applications.
2024
SRGS: Super-Resolution 3D Gaussian Splatting
Submitted to ACM Multimedia (ACM MM) Under review. 2026 Core A* CCF A
This work provides a novel framework for recovering high-quality 3D Gaussian Splatting representations from low-resolution multi-view captures. By integrating 2D super-resolution priors with cross-view geometric consistency, it supports more robust 3D reconstruction for real-world rendering, virtual reality, and immersive visualization applications.
SRGS: Super-Resolution 3D Gaussian Splatting
Submitted to ACM Multimedia (ACM MM) Under review. 2026 Core A* CCF A
This work provides a novel framework for recovering high-quality 3D Gaussian Splatting representations from low-resolution multi-view captures. By integrating 2D super-resolution priors with cross-view geometric consistency, it supports more robust 3D reconstruction for real-world rendering, virtual reality, and immersive visualization applications.
RT-NeRV: Rethinking Hybrid Neural Representations for Video via Residual Tokenization
Submitted to ACM Multimedia (ACM MM) Under review. 2026 Core A* CCF A
This work develops residual tokenization framework for efficient and high-quality neural video compression. By transmitting compact residual tokens instead of dense shallow features, it improves visual detail preservation while keeping the representation compact and practical for video storage and transmission.
RT-NeRV: Rethinking Hybrid Neural Representations for Video via Residual Tokenization
Submitted to ACM Multimedia (ACM MM) Under review. 2026 Core A* CCF A
This work develops residual tokenization framework for efficient and high-quality neural video compression. By transmitting compact residual tokens instead of dense shallow features, it improves visual detail preservation while keeping the representation compact and practical for video storage and transmission.
Privacy-preserving in medical image analysis: A review of methods and applications
The 25th International Conference on Parallel and Distributed Computing: Applications and Technologies (PDCAT) 2024
This work examines privacy-preserving technologies for medical image analysis. It supports the development of secure, scalable, and trustworthy medical AI systems that protect patient data while maintaining diagnostic utility.
Privacy-preserving in medical image analysis: A review of methods and applications
The 25th International Conference on Parallel and Distributed Computing: Applications and Technologies (PDCAT) 2024
This work examines privacy-preserving technologies for medical image analysis. It supports the development of secure, scalable, and trustworthy medical AI systems that protect patient data while maintaining diagnostic utility.
AC-UNet: Adaptive Connection UNet for White Matter Tract Segmentation Through Neural Architecture Search
IEEE 21th International Symposium on Biomedical Imaging (ISBI) 2024 ERA2010 Core A
This work develops AC-UNet for accurate white matter tract segmentation in diffusion MRI. It enables more accurate delineation of complex brain fiber pathways, and supports downstream applications in neurodegenerative disease analysis, brain injury assessment, and surgical planning.
AC-UNet: Adaptive Connection UNet for White Matter Tract Segmentation Through Neural Architecture Search
IEEE 21th International Symposium on Biomedical Imaging (ISBI) 2024 ERA2010 Core A
This work develops AC-UNet for accurate white matter tract segmentation in diffusion MRI. It enables more accurate delineation of complex brain fiber pathways, and supports downstream applications in neurodegenerative disease analysis, brain injury assessment, and surgical planning.
2023
The cell tracking challenge: 10 years of objective benchmarking
Nature Methods 2023 JCR Q1
This work contributes to the benchmarking infrastructure of bioimage analysis by systematically evaluating cell segmentation and tracking algorithms across diverse microscopy datasets. It supports the development of more robust, generalizable, and reusable methods for large-scale biological image analysis.
The cell tracking challenge: 10 years of objective benchmarking
Nature Methods 2023 JCR Q1
This work contributes to the benchmarking infrastructure of bioimage analysis by systematically evaluating cell segmentation and tracking algorithms across diverse microscopy datasets. It supports the development of more robust, generalizable, and reusable methods for large-scale biological image analysis.
A comparison of manual and automated neural architecture search for white matter tract segmentation
Scientific Reports 2023 JCR Q1
This work investigates deep network architectures for white matter tract segmentation in diffusion MRI, an important task for brain connectivity analysis, neurodegenerative disease studies, and surgical planning. It provides guidance for building more reliable brain tract segmentation models.
A comparison of manual and automated neural architecture search for white matter tract segmentation
Scientific Reports 2023 JCR Q1
This work investigates deep network architectures for white matter tract segmentation in diffusion MRI, an important task for brain connectivity analysis, neurodegenerative disease studies, and surgical planning. It provides guidance for building more reliable brain tract segmentation models.
FingerGAN: A Constrained Fingerprint Generation Scheme for Latent Fingerprint Enhancement
IEEE Transactions on Pattern Analysis and Machine Intelligence 2023 JCR Q1
This work introduces FingerGAN, a GAN based latent fingerprint enhancement method that directly optimizes fingerprint skeleton and minutia information. It improves degraded latent fingerprint quality and supports more accurate forensic fingerprint identification.
FingerGAN: A Constrained Fingerprint Generation Scheme for Latent Fingerprint Enhancement
IEEE Transactions on Pattern Analysis and Machine Intelligence 2023 JCR Q1
This work introduces FingerGAN, a GAN based latent fingerprint enhancement method that directly optimizes fingerprint skeleton and minutia information. It improves degraded latent fingerprint quality and supports more accurate forensic fingerprint identification.
2022
Deep learning in diverse intelligent sensor based systems
Sensors 2022
This work reviews deep learning methods and applications across diverse intelligent sensor=based systems. It connects fundamental models with practical sensor driven domains, providing a broad reference for developing intelligent sensing, perception, and decision making systems.
Deep learning in diverse intelligent sensor based systems
Sensors 2022
This work reviews deep learning methods and applications across diverse intelligent sensor=based systems. It connects fundamental models with practical sensor driven domains, providing a broad reference for developing intelligent sensing, perception, and decision making systems.
A compound loss function with shape aware weight map for microscopy cell segmentation
IEEE Transactions on Medical Imaging 2022 JCR Q1
This work develops a shape-aware compound loss function for microscopy cell segmentation. It improves segmentation accuracy by guiding deep models to better capture cell boundaries, subtle cell structures, and challenging regions with missing or imbalanced annotations.
A compound loss function with shape aware weight map for microscopy cell segmentation
IEEE Transactions on Medical Imaging 2022 JCR Q1
This work develops a shape-aware compound loss function for microscopy cell segmentation. It improves segmentation accuracy by guiding deep models to better capture cell boundaries, subtle cell structures, and challenging regions with missing or imbalanced annotations.
A Novel Length-Flexible Lightweight Cancelable Fingerprint Template for Privacy-Preserving Authentication Systems in Resource-Constrained IoT Applications
IEEE Internet of Things Journal 2022 JCR Q1
This work introduces PowerFDNet for detecting stealthy false data injection attacks in AC-model smart grids. It learns both spatial relationships among buses and transmission lines and temporal patterns in measurement sequences, enabling more accurate and deployable cyber-attack detection for power transmission systems.
A Novel Length-Flexible Lightweight Cancelable Fingerprint Template for Privacy-Preserving Authentication Systems in Resource-Constrained IoT Applications
IEEE Internet of Things Journal 2022 JCR Q1
This work introduces PowerFDNet for detecting stealthy false data injection attacks in AC-model smart grids. It learns both spatial relationships among buses and transmission lines and temporal patterns in measurement sequences, enabling more accurate and deployable cyber-attack detection for power transmission systems.
PowerFDNet: Deep Learning-Based Stealthy False Data Injection Attack Detection for AC-model Transmission Systems
IEEE Open Journal of the Computer Society 2022 JCR Q1
This work introduces PowerFDNet for detecting stealthy false data injection attacks in AC-model smart grids. It learns both spatial relationships among buses and transmission lines and temporal patterns in measurement sequences, enabling more accurate and deployable cyber-attack detection for power transmission systems.
PowerFDNet: Deep Learning-Based Stealthy False Data Injection Attack Detection for AC-model Transmission Systems
IEEE Open Journal of the Computer Society 2022 JCR Q1
This work introduces PowerFDNet for detecting stealthy false data injection attacks in AC-model smart grids. It learns both spatial relationships among buses and transmission lines and temporal patterns in measurement sequences, enabling more accurate and deployable cyber-attack detection for power transmission systems.
Representation Learning and Pattern Recognition in Cognitive Biometrics: A Survey
Sensors 2022
This work reviews representation learning and pattern recognition methods for cognitive biometrics, covering major biosignals such as EEG, ECG, PPG, EOG, EMG, and EDA. It provides a structured overview of signal acquisition, preprocessing, feature learning, recognition models, and future directions.
Representation Learning and Pattern Recognition in Cognitive Biometrics: A Survey
Sensors 2022
This work reviews representation learning and pattern recognition methods for cognitive biometrics, covering major biosignals such as EEG, ECG, PPG, EOG, EMG, and EDA. It provides a structured overview of signal acquisition, preprocessing, feature learning, recognition models, and future directions.
2021
Automatic improvement of deep learning-based cell segmentation in time-lapse microscopy by neural architecture search
Bioinformatics 2021 JCR Q1
This work develops an automated neural architecture search method for live cell segmentation in time-lapse microscopy. It designs task-specific networks that exploit both spatial and temporal information, achieving robust performance across diverse microscopy cell datasets.
Automatic improvement of deep learning-based cell segmentation in time-lapse microscopy by neural architecture search
Bioinformatics 2021 JCR Q1
This work develops an automated neural architecture search method for live cell segmentation in time-lapse microscopy. It designs task-specific networks that exploit both spatial and temporal information, achieving robust performance across diverse microscopy cell datasets.
A survey on 2D and 3D contactless fingerprint biometrics: A taxonomy, review, and future directions
IEEE Open Journal of the Computer Society 2021 JCR Q1
This work reviews 2D and 3D contactless fingerprint biometics. It provide a taxinomy if capture techniques, preprocessing methods, feature extraction approaches and comparison strategies, offering guided overview of this emerging biometirc field.
A survey on 2D and 3D contactless fingerprint biometrics: A taxonomy, review, and future directions
IEEE Open Journal of the Computer Society 2021 JCR Q1
This work reviews 2D and 3D contactless fingerprint biometics. It provide a taxinomy if capture techniques, preprocessing methods, feature extraction approaches and comparison strategies, offering guided overview of this emerging biometirc field.
A subgrid-oriented privacy-preserving microservice framework based on deep neural network for false data injection attack detection in smart grids
IEEE Transactions on Industrial Informatics 2021 JCR Q1
This work develops a privacy-preserving microservice framework for detecting false data injection attacks in smart grids. It performs local spatial-temporal learning within subgrids and collaboratively detects attacks through subgrid-level representations, supporting accurate, low-latency, and privacy-aware smart grid security.
A subgrid-oriented privacy-preserving microservice framework based on deep neural network for false data injection attack detection in smart grids
IEEE Transactions on Industrial Informatics 2021 JCR Q1
This work develops a privacy-preserving microservice framework for detecting false data injection attacks in smart grids. It performs local spatial-temporal learning within subgrids and collaboratively detects attacks through subgrid-level representations, supporting accurate, low-latency, and privacy-aware smart grid security.
A comprehensive survey of privacy-preserving federated learning: A taxonomy, review, and future directions
ACM Computing Surveys (CSUR) 2021 JCR Q1
This work reviews privacy-preserving federated learning through a 5W-scenario-based taxonomy. It analyzes privacy leakage risks in federated learning and summarizes key protection mechanisms, providing a structured reference for secure and trustworthy distributed machine learning.
A comprehensive survey of privacy-preserving federated learning: A taxonomy, review, and future directions
ACM Computing Surveys (CSUR) 2021 JCR Q1
This work reviews privacy-preserving federated learning through a 5W-scenario-based taxonomy. It analyzes privacy leakage risks in federated learning and summarizes key protection mechanisms, providing a structured reference for secure and trustworthy distributed machine learning.
Neural architecture search for microscopy cell segmentation
International Workshop on Machine Learning in Medical Imaging 2020
This work applies neural architecture search to microscopy cell segmentation. It automatically optimizes U-Net and UNet++ based architectures using task-specific network blocks, improving segmentation performance while reducing manual network design effort.
Neural architecture search for microscopy cell segmentation
International Workshop on Machine Learning in Medical Imaging 2020
This work applies neural architecture search to microscopy cell segmentation. It automatically optimizes U-Net and UNet++ based architectures using task-specific network blocks, improving segmentation performance while reducing manual network design effort.
Before 2020
3D fingerprint recognition based on ridge-valley-guided 3D reconstruction and 3D topology polymer feature extraction
IEEE Transactions on Pattern Analysis and Machine Intelligence 2019 JCR Q1
This work develops a real-time 3D fingerprint recognition method that combines ridge-valley-guided reconstruction with 3D topology polymer features. It improves both reconstruction quality and recognition accuracy, supporting practical and contactless biometric authentication.
3D fingerprint recognition based on ridge-valley-guided 3D reconstruction and 3D topology polymer feature extraction
IEEE Transactions on Pattern Analysis and Machine Intelligence 2019 JCR Q1
This work develops a real-time 3D fingerprint recognition method that combines ridge-valley-guided reconstruction with 3D topology polymer features. It improves both reconstruction quality and recognition accuracy, supporting practical and contactless biometric authentication.
Contactless fingerprint recognition based on global minutia topology and loose genetic algorithm
IEEE Transactions on Information Forensics and Security 2019 JCR Q1
This work applies neural architecture search to microscopy cell segmentation. It automatically optimizes U-Net and UNet++ based architectures using task-specific network blocks, improving segmentation performance while reducing manual network design effort.
Contactless fingerprint recognition based on global minutia topology and loose genetic algorithm
IEEE Transactions on Information Forensics and Security 2019 JCR Q1
This work applies neural architecture search to microscopy cell segmentation. It automatically optimizes U-Net and UNet++ based architectures using task-specific network blocks, improving segmentation performance while reducing manual network design effort.
Robust fingerprint matching based on convolutional neural networks
International Conference on Mobile Networks and Management 2017
This work develops a CNN based fingerprint matching method that learns identity similarities directly from raw fingerprint pairs. It improves matching robustness, especially for incomplete and partial fingerprints.
Robust fingerprint matching based on convolutional neural networks
International Conference on Mobile Networks and Management 2017
This work develops a CNN based fingerprint matching method that learns identity similarities directly from raw fingerprint pairs. It improves matching robustness, especially for incomplete and partial fingerprints.
A robust contactless fingerprint enhancement algorithm
International Conference on Mobile Networks and Management 2017
This work develops a robust contactless fingerprint enhancement algorithm. It improves unclear ridge-valley patterns using preprocessing, sinusoidal-shaped filtering, and score filtering, supporting more reliable minutiae extraction and fingerprint verification.
A robust contactless fingerprint enhancement algorithm
International Conference on Mobile Networks and Management 2017
This work develops a robust contactless fingerprint enhancement algorithm. It improves unclear ridge-valley patterns using preprocessing, sinusoidal-shaped filtering, and score filtering, supporting more reliable minutiae extraction and fingerprint verification.
Latent fingerprint segmentation based on convolutional neural networks
IEEE Workshop on Information Forensics and Security (WIFS) 2017
This work is the first to introduce CNN into the field. It formulate the segmentation problem as a classification system, achieving robust latent fingerprint segmentation.
Latent fingerprint segmentation based on convolutional neural networks
IEEE Workshop on Information Forensics and Security (WIFS) 2017
This work is the first to introduce CNN into the field. It formulate the segmentation problem as a classification system, achieving robust latent fingerprint segmentation.
A robust multi-constrained model for fingerprint orientation field construction
IEEE 11th Conference on Industrial Electronics and Applications (ICIEA) 2016
This work develops a robust model for fingerprint orientation field construction. It combines multiple regularization constraints to preserve orientation structure, suppress noise, and improve the reliability of fingerprint indexing.
A robust multi-constrained model for fingerprint orientation field construction
IEEE 11th Conference on Industrial Electronics and Applications (ICIEA) 2016
This work develops a robust model for fingerprint orientation field construction. It combines multiple regularization constraints to preserve orientation structure, suppress noise, and improve the reliability of fingerprint indexing.
Video super-resolution using an adaptive superpixel-guided auto-regressive model
Pattern Recognition 2016 JCR Q1
This work develops an adaptive superpixel-guided model for video super-resolution. It combines key-frame selection, optical-flow-based temporal alignment, and superpixel-guided spatial modeling to recover sharper details and improve video quality efficiently.
Video super-resolution using an adaptive superpixel-guided auto-regressive model
Pattern Recognition 2016 JCR Q1
This work develops an adaptive superpixel-guided model for video super-resolution. It combines key-frame selection, optical-flow-based temporal alignment, and superpixel-guided spatial modeling to recover sharper details and improve video quality efficiently.
Video super-resolution based on automatic key-frame selection and feature-guided variational optical flow
Signal Processing: Image Communication 2014
TThis work develops a video super-resolution method for enhancing low-resolution video sequences. It improves visual quality while reducing common artifacts, supporting scenarios where video quality is limited by camera resolution, bandwidth, or storage constraints.
Video super-resolution based on automatic key-frame selection and feature-guided variational optical flow
Signal Processing: Image Communication 2014
TThis work develops a video super-resolution method for enhancing low-resolution video sequences. It improves visual quality while reducing common artifacts, supporting scenarios where video quality is limited by camera resolution, bandwidth, or storage constraints.