Jiabin Yu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Jiabin Yu has a total of 71 co-authors. Here are the top co-authors by number of shared publications:
- Dongping Zhang
China Jiliang University
4 shared publications - Binggang Xiao
China Jiliang University, College of Information Engineering, Key Laboratory of Electromagnetic Wave Information Technology and Metrology of Zhejiang Province, Hangzhou, China
3 shared publications - Dong Xu
Department of Diagnostic Ultrasound Imaging & Interventional Therapy, Hangzhou Institute of Medicine (HIM), Zhejiang Cancer Hospital, Chinese Academy of Sciences, Hangzhou, China
3 shared publications Hongmei Mi
3 shared publications- Hongmei Mi
Key Laboratory of Electromagnetic Wave Information Technology and Metrology of Zhejiang Province College of Information Engineering China Jiliang University Hangzhou China
3 shared publications - Li Yang
College of Information Engineering, China Jiliang University, Hangzhou 310018, China
3 shared publications
Latest Publications
- 19 November 2026
- Published by Elsevier in Biomedical Signal Processing and Control
- Vol. 131
C2F-Net: Coupled and Cross-Layer Feature Enhancement for Thyroid Ultrasound Image Segmentation
- 19 November 2026
- Published by Elsevier in Biomedical Signal Processing and Control
- Vol. 131
C2F-Net: Coupled and Cross-Layer Feature Enhancement for Thyroid Ultrasound Image Segmentation
- 22 October 2026
- Published by Wiley in Annalen der Physik
- Vol. 538 (10) , e70320
Quad‐Band Terahertz Electromagnetically Induced Transparency‐Like Metasurface Sensor With Geometry‐Constrained Inversion
This study proposes a quad‐band terahertz electromagnetically induced transparency (EIT)‐like metasurface sensor with geometry‐constrained parameter inversion. The device consists of three concentric ring resonators and a bottom U‐shaped resonator. A gap in the inner ring breaks symmetry, while cooperative bright–dark and dark–dark mode coupling produces four transparency windows at 0.49, 0.59, 0.76, and 0.86 THz. Refractive‐index sensing simulations yield a maximum sensitivity of 269.8 GHz/RIU (RIU, refractive index unit) and a maximum Q factor of 66.4. Further experiments with antibiotic samples are required to validate practical detection performance. An improved multiscale temporal convolutional network (Improved‐TCN) captures multiscale local features and cross‐band correlations in the multimode‐coupled spectra. By integrating multiscale dilated convolutions and self‐attention, it predicts a geometrically feasible candidate set of five parameters from 201‐point linear transmission amplitude spectra. Across five independent runs, the full model achieves a standardized mean squared error (MSE) of 0.028 ± 0.001 and a coefficient of determination (R 2 ) of 0.969 ± 0.003, demonstrating stable performance on the held‐out test set. The method provides a practical reference for rapid, geometrically feasible parameter inversion of multiband terahertz sensors with predefined topologies.
- 22 October 2026
- Published by Wiley in Annalen der Physik
- Vol. 538 (10) , e70320
Quad‐Band Terahertz Electromagnetically Induced Transparency‐Like Metasurface Sensor With Geometry‐Constrained Inversion
This study proposes a quad‐band terahertz electromagnetically induced transparency (EIT)‐like metasurface sensor with geometry‐constrained parameter inversion. The device consists of three concentric ring resonators and a bottom U‐shaped resonator. A gap in the inner ring breaks symmetry, while cooperative bright–dark and dark–dark mode coupling produces four transparency windows at 0.49, 0.59, 0.76, and 0.86 THz. Refractive‐index sensing simulations yield a maximum sensitivity of 269.8 GHz/RIU (RIU, refractive index unit) and a maximum Q factor of 66.4. Further experiments with antibiotic samples are required to validate practical detection performance. An improved multiscale temporal convolutional network (Improved‐TCN) captures multiscale local features and cross‐band correlations in the multimode‐coupled spectra. By integrating multiscale dilated convolutions and self‐attention, it predicts a geometrically feasible candidate set of five parameters from 201‐point linear transmission amplitude spectra. Across five independent runs, the full model achieves a standardized mean squared error (MSE) of 0.028 ± 0.001 and a coefficient of determination (R 2 ) of 0.969 ± 0.003, demonstrating stable performance on the held‐out test set. The method provides a practical reference for rapid, geometrically feasible parameter inversion of multiband terahertz sensors with predefined topologies.
- 25 July 2026
- Published by Elsevier in Computer Methods and Programs in Biomedicine
- Vol. 286, 109569
BMR-Restormer: A transformer-based blind inpainting network for ultrasound marker removal
- 25 July 2026
- Published by Elsevier in Computer Methods and Programs in Biomedicine
- Vol. 286, 109569
BMR-Restormer: A transformer-based blind inpainting network for ultrasound marker removal
- Open Access
- 4 May 2026
- Published by Institution of Engineering and Technology (IET) in CAAI Transactions on Intelligence Technology
- Vol. 11 (3) , 900-919
A Semantic Segmentation Network for Colorectal Polyp Images With Progressive Fusion of Dual‐Branch Features
Accurate segmentation of colorectal polyps is essential for early colorectal cancer screening, yet remains challenging due to weak foreground–background contrast, disrupted boundaries caused by specular reflections and intestinal folds, and pronounced scale variation among polyps. These factors make it difficult for existing methods to jointly preserve fine boundary details and robust global semantic context. To address these task‐specific challenges, we propose a Dual‐branch Feature Progressive Fusion Network (DFPF‐Net) for colorectal polyp segmentation. DFPF‐Net adopts a dual‐encoder architecture that integrates a CNN‐based encoder for local and boundary‐sensitive representation for global semantic modelling. A boundary‐aware branch equipped with stacked Inversely Perceive Information Layers (IPILs) enhances ambiguous and fragmented contours, while the semantic branch incorporates Misalignment Fusion Modules (MFMs) and a Misaligned Single‐layer Reinforcement Module (MSRM) to alleviate semantic misalignment and insufficient cross‐scale interaction. Furthermore, a Perceptual Information Fusion Module (PIFM) enables effective semantic–boundary collaboration, and a Multi‐level Residual Decoding Module (MRDM) progressively reconstructs structurally consistent segmentation outputs. Extensive experiments on multiple public colonoscopy datasets demonstrate that DFPF‐Net achieves competitive and robust segmentation performance. In particular, on the challenging ETIS dataset, DFPF‐Net attains 0.785 mDice and 0.704 mIoU, indicating its capability in handling complex structures and ambiguous boundaries in colorectal polyp segmentation.
- 4 May 2026
- Published by Institution of Engineering and Technology (IET) in CAAI Transactions on Intelligence Technology
- Vol. 11 (3) , 900-919
Open AccessA Semantic Segmentation Network for Colorectal Polyp Images With Progressive Fusion of Dual‐Branch Features
Accurate segmentation of colorectal polyps is essential for early colorectal cancer screening, yet remains challenging due to weak foreground–background contrast, disrupted boundaries caused by specular reflections and intestinal folds, and pronounced scale variation among polyps. These factors make it difficult for existing methods to jointly preserve fine boundary details and robust global semantic context. To address these task‐specific challenges, we propose a Dual‐branch Feature Progressive Fusion Network (DFPF‐Net) for colorectal polyp segmentation. DFPF‐Net adopts a dual‐encoder architecture that integrates a CNN‐based encoder for local and boundary‐sensitive representation for global semantic modelling. A boundary‐aware branch equipped with stacked Inversely Perceive Information Layers (IPILs) enhances ambiguous and fragmented contours, while the semantic branch incorporates Misalignment Fusion Modules (MFMs) and a Misaligned Single‐layer Reinforcement Module (MSRM) to alleviate semantic misalignment and insufficient cross‐scale interaction. Furthermore, a Perceptual Information Fusion Module (PIFM) enables effective semantic–boundary collaboration, and a Multi‐level Residual Decoding Module (MRDM) progressively reconstructs structurally consistent segmentation outputs. Extensive experiments on multiple public colonoscopy datasets demonstrate that DFPF‐Net achieves competitive and robust segmentation performance. In particular, on the challenging ETIS dataset, DFPF‐Net attains 0.785 mDice and 0.704 mIoU, indicating its capability in handling complex structures and ambiguous boundaries in colorectal polyp segmentation.
- Open Access
- 1 May 2025
- Published by SAGE Publications in Digital Health
- Vol. 11
Weakly supervised deep learning for multimodal MRI-TRUS registration: Toward assisting prostate biopsy guidance
Objective: Accurate magnetic resonance imaging-transrectal ultrasound (MRI-TRUS) registration is essential for improving prostate cancer detection and biopsy guidance. This study presents a weakly supervised learning framework to enhance image alignment while minimizing reliance on extensive labeled data. Methods: We developed a two-stage weakly supervised framework integrating an attention-enhanced U-Net for prostate segmentation and a residual-enhanced registration network (RERN) for MRI-TRUS alignment. The segmentation model was trained and evaluated on public MRI datasets (MSD Prostate, Promise12, µ-RegPro) and TRUS datasets (µ-RegPro), with additional testing on a clinical cohort of 32 MRI-TRUS pairs. Performance was assessed using dice similarity coefficient (DSC), accuracy (Acc), precision (Pre), recall (Rec), and the 95th percentile Hausdorff distance (HD95). The registration model was trained and tested using MRI-TRUS pairs from µ-RegPro and the clinical cohort, with performance evaluated based on HD95 and target registration error (TRE). Additionally, clinical validation included PI-RADS-like grading, Likert confidence scoring, and receiver operating characteristic (ROC) analysis to assess diagnostic impact. Results: The segmentation model demonstrated high accuracy (DSC: MRI 0.9154, TRUS 0.9384) and strong generalizability to clinical data (MRI 0.9010, TRUS 0.9173). The registration model achieved robust performance, with HD95 of 10.18 mm on public datasets and 11.18 mm on clinical data, and TRE below 8.64 mm. Clinical validation confirmed that registered MRI images preserved diagnostic integrity, as no significant differences were observed in radiologists’ diagnostic performance (AUC: junior 0.706, senior 0.781, p > 0.05). Moreover, registered images enhanced diagnostic confidence among senior radiologists (Likert score: 3.032 vs. 3.548, p = 0.015), highlighting their potential to support expert-level decision-making. Conclusions: The proposed weakly supervised framework achieves high-precision MRI-TRUS registration with minimal annotation, ensuring strong generalizability and clinical applicability.
- 1 May 2025
- Published by SAGE Publications in Digital Health
- Vol. 11
Open AccessWeakly supervised deep learning for multimodal MRI-TRUS registration: Toward assisting prostate biopsy guidance
Objective: Accurate magnetic resonance imaging-transrectal ultrasound (MRI-TRUS) registration is essential for improving prostate cancer detection and biopsy guidance. This study presents a weakly supervised learning framework to enhance image alignment while minimizing reliance on extensive labeled data. Methods: We developed a two-stage weakly supervised framework integrating an attention-enhanced U-Net for prostate segmentation and a residual-enhanced registration network (RERN) for MRI-TRUS alignment. The segmentation model was trained and evaluated on public MRI datasets (MSD Prostate, Promise12, µ-RegPro) and TRUS datasets (µ-RegPro), with additional testing on a clinical cohort of 32 MRI-TRUS pairs. Performance was assessed using dice similarity coefficient (DSC), accuracy (Acc), precision (Pre), recall (Rec), and the 95th percentile Hausdorff distance (HD95). The registration model was trained and tested using MRI-TRUS pairs from µ-RegPro and the clinical cohort, with performance evaluated based on HD95 and target registration error (TRE). Additionally, clinical validation included PI-RADS-like grading, Likert confidence scoring, and receiver operating characteristic (ROC) analysis to assess diagnostic impact. Results: The segmentation model demonstrated high accuracy (DSC: MRI 0.9154, TRUS 0.9384) and strong generalizability to clinical data (MRI 0.9010, TRUS 0.9173). The registration model achieved robust performance, with HD95 of 10.18 mm on public datasets and 11.18 mm on clinical data, and TRE below 8.64 mm. Clinical validation confirmed that registered MRI images preserved diagnostic integrity, as no significant differences were observed in radiologists’ diagnostic performance (AUC: junior 0.706, senior 0.781, p > 0.05). Moreover, registered images enhanced diagnostic confidence among senior radiologists (Likert score: 3.032 vs. 3.548, p = 0.015), highlighting their potential to support expert-level decision-making. Conclusions: The proposed weakly supervised framework achieves high-precision MRI-TRUS registration with minimal annotation, ensuring strong generalizability and clinical applicability.
- 1 May 2025
- Published by Radiological Society of North America (RSNA) in Radiology: Artificial Intelligence
- Vol. 7 (3) , e240370
Pseudo–Contrast-enhanced US via Enhanced Generative Adversarial Networks for Evaluating Tumor Ablation Efficacy
The pseudo–contrast-enhanced US (CEUS) method, based on enhanced generative adversarial networks, demonstrated high similarity to real CEUS in evaluating tumor ablation effectiveness.
- 1 May 2025
- Published by Radiological Society of North America (RSNA) in Radiology: Artificial Intelligence
- Vol. 7 (3) , e240370
Pseudo–Contrast-enhanced US via Enhanced Generative Adversarial Networks for Evaluating Tumor Ablation Efficacy
The pseudo–contrast-enhanced US (CEUS) method, based on enhanced generative adversarial networks, demonstrated high similarity to real CEUS in evaluating tumor ablation effectiveness.
- 8 March 2025
- Published by Elsevier in Applied Soft Computing
- Vol. 174
Graph convolutional networks with multi-scale dynamics for traffic speed forecasting
- 8 March 2025
- Published by Elsevier in Applied Soft Computing
- Vol. 174
Graph convolutional networks with multi-scale dynamics for traffic speed forecasting
- 12 November 2024
- Published by Elsevier in Computers & Security
- Vol. 149
PdGAT-ID: An intrusion detection method for industrial control systems based on periodic extraction and spatiotemporal graph attention
- 12 November 2024
- Published by Elsevier in Computers & Security
- Vol. 149
PdGAT-ID: An intrusion detection method for industrial control systems based on periodic extraction and spatiotemporal graph attention