Publications

Search or filter publications

Search publications Search

Filter by type:

Filter by publication type

• Book
• Book chapter
• Conference paper
• Journal article
• Other
• Patent
• Poster
• Report
• Scholarly edition
• Software
• Thesis dissertation
• Working paper

Filter by year:

Filter from 2026202520242023202220212020201920182017201620152014201320122011201020092008200720062005200420032002 to Filter to 2026202520242023202220212020201920182017201620152014201320122011201020092008200720062005200420032002

---

Search results

• Journal article
  Zhou T, Li M, Ruan S, Luo T, Jiang B, Zhu J, Ma P, Yang D, Yang Get al., 2026,

  A reliable framework for brain tumor segmentation via multi-modal fusion and uncertainty modeling

  , Information Fusion, Vol: 129, ISSN: 1566-2535

  Accurate brain tumor segmentation from MRI scans is critical for effective diagnosis and treatment planning. Recent advances in deep learning have significantly improved brain tumor segmentation performance. However, these models still face challenges in clinical adoption due to their inherent uncertainties and potential for errors. In this paper, we propose a novel MR brain tumor segmentation approach that integrates multi-modal data fusion and uncertainty quantification to improve the accuracy and reliability of brain tumor segmentation. Recognizing that each MR modality contributes unique insights into the tumor’s characteristics, we propose a novel modality-aware guidance by explicitly categorizing the modalities into ”teacher” (FLAIR and T1c) and ”student” (T2 and T1) groups. Since the teacher modalities are the most informative modalities for identifying brain tumors, we propose a multi-modal teacher-student fusion strategy. This strategy leverages the teacher modalities to guide the student modalities in both spatial and channel feature representation aspects. To address prediction reliability, we employ Monte Carlo dropout during training to generate multiple uncertainty estimates. Additionally, we develop a novel uncertainty-aware loss function that optimizes segmentation accuracy while quantifying the uncertainty in predictions. Experimental results conducted on three BraTS datasets demonstrate the effectiveness of the proposed components and the superior performance compared to the state-of-the-art methods, highlighting their potential for clinical application.

  • Abstract
  • Cite
• Journal article
  Hasan MK, Yang G, Yap CH, 2026,

  An efficient, scalable, and adaptable plug-and-play temporal attention module for motion-guided cardiac segmentation with sparse temporal labels

  , Medical Image Analysis, Vol: 110, Pages: 103981-103981, ISSN: 1361-8415
  • Publisher Web Link
  • Cite
• Journal article
  Liao Y, Zheng Y, Zhu J, Chen Y, Gao F, Feng Y, Yang W, Yang G, Lai X, Li Pet al., 2026,

  Self-attention-based mixture-of-experts framework for non-invasive prediction of MGMT promoter methylation in glioblastoma using multi-modal MRI

  , Displays, Vol: 92, ISSN: 0141-9382

  Glioblastoma (GBM) is an aggressive brain tumor associated with poor prognosis and limited treatment options. The methylation status of the O6-methylguanine-DNA methyltransferase (MGMT) promoter is a critical biomarker for predicting the efficacy of temozolomide chemotherapy in GBM patients. However, current methods for determining MGMT promoter methylation, including invasive and costly techniques, hinder their widespread clinical application. In this study, we propose a novel non-invasive deep learning framework based on a Mixture-of-Experts (MoE) architecture for predicting MGMT promoter methylation status using multi-modal magnetic resonance imaging (MRI) data. Our MoE model incorporates modality-specific expert networks built on the ResNet18 architecture, with a self-attention-based gating mechanism that dynamically selects and integrates the most relevant features across MRI modalities (T1-weighted, contrast-enhanced T1, T2-weighted, and fluid-attenuated inversion recovery). We evaluate the proposed framework on the BraTS2021 and TCGA-GBM datasets, showing superior performance compared to conventional deep learning models in terms of accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC). Furthermore, Grad-CAM visualizations provide enhanced interpretability by highlighting biologically relevant regions in the tumor and peritumoral areas that influence model predictions. The proposed framework represents a promising tool for integrating imaging biomarkers into precision oncology workflows, offering a scalable, cost-effective, and interpretable solution for non-invasive MGMT methylation prediction in GBM.

  • Abstract
  • Cite
• Journal article
  Cheng CW, Huang J, Zhang Y, Yang G, Schönlieb CB, Aviles-Rivero AIet al., 2026,

  Mamba neural operator: Who wins? transformers vs. state-space models for PDEs

  , Journal of Computational Physics, Vol: 548, ISSN: 0021-9991

  Partial differential equations (PDEs) are widely used to model complex physical systems, but solving them efficiently remains a significant challenge. Recently, Transformers have emerged as the preferred architecture for PDEs due to their ability to capture intricate dependencies. However, they struggle with representing continuous dynamics and long-range interactions. To overcome these limitations, we introduce the Mamba Neural Operator (MNO), a novel framework that enhances neural operator-based techniques for solving PDEs. MNO establishes a formal theoretical connection between structured state-space models (SSMs) and neural operators, offering a unified structure that can adapt to diverse architectures, including Transformer-based models. By leveraging the structured design of SSMs, MNO captures long-range dependencies and continuous dynamics more effectively than traditional Transformers. Through extensive analysis, we show that MNO significantly boosts the expressive power and accuracy of neural operators, making it not just a complement but a superior framework for PDE-related tasks, bridging the gap between efficient representation and accurate solution approximation.

  • Abstract
  • Cite
  • Citations: 1
• Journal article
  Ma X, Tao Y, Zhang Z, Zhang Y, Wang X, Zhang S, Ji Z, Zhang Y, Chen Q, Yang Get al., 2026,

  Test-time generative augmentation for medical image segmentation

  , MEDICAL IMAGE ANALYSIS, Vol: 109, ISSN: 1361-8415
  • Cite
• Journal article
  Jing P, Lee K, Zhang Z, Zhou H, Yuan Z, Gao Z, Zhu L, Papanastasiou G, Fang Y, Yang Get al., 2026,

  Reason like a radiologist: Chain-of-thought and reinforcement learning for verifiable report

  , MEDICAL IMAGE ANALYSIS, Vol: 109, ISSN: 1361-8415
  • Cite
• Journal article
  Li K, Xiao X, Zhong Z, Yang Get al., 2026,

  Accurate and generalizable protein-ligand binding affinity prediction with geometric deep learning

  , IEEE Open Journal of Engineering in Medicine and Biology, ISSN: 2644-1276

  Goal: Protein-ligand binding complexes are ubiquitous and essential to life. Protein-ligand binding affinity prediction (PLA) quantifies the binding strength between ligands and proteins, providing crucial insights for discovering and designing potential candidate ligands. While recent advances have been made in predicting protein-ligand complex structures, existing algorithms for interaction and affinity prediction suffer from a sharp decline in performance when handling ligands bound with novel unseen proteins. Methods: We propose IPBind, a geometric deep learning-based computational method, enabling robust predictions by leveraging interatomic potential between complex’s bound and unbound status. Results: Experimental results on widely used binding affinity prediction benchmarks demonstrate the effectiveness and universality of IPBind. Meanwhile, it provids atom-level insights into prediction. Conclusions: This work highlight the advantage of leveraging machine learning interatomic potential for predicting protein-ligand binding affinity. Index Terms—Deep learning, drug discovery, physics-informed neural networks, protein-ligand binding affinity prediction. Impact Statement–This study extends state-of-the-art deep learning algorithms to applications in protein-ligand binding affinity prediction. This study has implications for enhancing the generalization capability of protein-ligand interactions prediction methods by interatomic potential modeling.

  • Abstract
  • Cite
• Journal article
  Ma J, Jiang M, Fang X, Chen J, Wang Y, Yang Get al., 2026,

  Hybrid aggregation strategy with double inverted residual blocks for lightweight salient object detection

  , NEURAL NETWORKS, Vol: 194, ISSN: 0893-6080
  • Cite
• Journal article
  Khalique Z, Scott AD, Ferreira PF, Molto M, Nielles-Vallespin S, Pennell DJet al., 2026,

  Diffusion Tensor CMR Assessment of the Microstructural Response to Dobutamine Stress in Health and Comparison With Patients With Recovered Dilated Cardiomyopathy.

  , Circ Cardiovasc Imaging, Vol: 19

  BACKGROUND: Contractile reserve assessment assesses myocardial performance and prognosis. The microstructural mechanisms that facilitate increased cardiac function have not been described, but can be studied using diffusion tensor cardiovascular magnetic resonance. Resting microstructural contractile function is characterized by reorientation of aggregated cardiomyocytes (sheetlets) from wall-parallel in diastole to a more wall-perpendicular configuration in systole, with the diffusion tensor cardiovascular magnetic resonance parameter E2A defining their orientation, and sheetlet mobility defining the angle through which they rotate. We used diffusion tensor cardiovascular magnetic resonance to identify the microstructural response to dobutamine stress in healthy volunteers and then compared with patients with recovered dilated cardiomyopathy (rDCM). METHODS: In this first-of-its-kind prospective observational study, 20 healthy volunteers and 32 patients with rDCM underwent diffusion tensor cardiovascular magnetic resonance at rest, during dobutamine, and on recovery. RESULTS: In healthy volunteers, both diastolic and systolic E2A increased with dobutamine stress (13±3° to 17±5°; P<0.001 and 59±11° to 65±7°; P=0.002). Sheetlet mobility remained unchanged (45±11° to 49±10°; P=0.19), but biphasic mean E2A increased (36±6° to 41±4°; P<0.001). In rDCM, diastolic E2A at rest was higher than in healthy volunteers (20±8° versus 13±3°, P<0.001), and sheetlet mobility was reduced (34±12° versus 45±11°; P<0.001). During dobutamine stress, rDCM diastolic and systolic E2A increased compared with rest (20±8° to 24±10°; P=0.001 and 54±13° to 63±11°; P=0.005). However, sheetlet mobility in patients with rDCM failed to increase with dobutamine to healthy levels (39±13° versus 49±

  • Abstract
  • Author Web Link
  • Cite
• Journal article
  Zhang S, Nan Y, Fang Y, Wang S, Liu Y, Papanastasiou G, Gao Z, Li S, Walsh S, Yang Get al., 2026,

  Dynamical multi-order responses and global semantic-infused adversarial learning: A robust airway segmentation method

  , MEDICAL IMAGE ANALYSIS, Vol: 108, ISSN: 1361-8415
  • Cite

This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.