Zhibin He
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Zhibin He has a total of 45 co-authors. Here are the top co-authors by number of shared publications:
- Yixuan Yuan
The Chinese University of Hong Kong
7 shared publications - Tuo Zhang
School of Automation Northwestern Polytechnical University Xi'an China
6 shared publications Wuyang Li
The Chinese University of Hong Kong
3 shared publications- Zhihao Peng
The Chinese University of Hong Kong
3 shared publications Junwei Han
School of Automation, Northwestern Polytechnical University, China
2 shared publications- Junwei Han
School of Artificial Intelligence, Chongqing University of Posts and Telecommunications, Chongqing, China
2 shared publications
Latest Publications
- 1 July 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 45 (7) , 3665-3676
Fiber HGNN: Heterogeneous Graph Neural Network for Fiber Tract Segmentation
Fiber tract segmentation is crucial for clinical applications such as brain function interpretation and surgical planning. Existing methods typically adopt either a cortical-parcellation-based or fiber clustering approach, but fail to simultaneously integrate heterogeneous information (e.g., streamline shape, point position, anatomical priors). In this work, we propose Fiber HGNN, a novel heterogeneous graph neural network that explicitly models and integrates heterogeneous information of fibers for accurate fiber tract segmentation. We construct a heterogeneous graph comprising three types of nodes: streamline, fiber keypoint and anatomical region. Specifically, fiber keypoints are representative points sampled along each streamline to characterize local geometric features, while anatomical regions provide contextual priors derived from brain atlas. This design enables the network to jointly capture the complementary information of streamline shape, local geometry, and anatomical priors, thus facilitating the learning of more discriminative feature representations. To further leverage implicit anatomical connectivity, we design a Metapath-guided Heterogeneous Information Aggregation (MHIA) network. By analyzing the spatial relationships between streamline keypoints and anatomical regions, the heterogeneous graph is decomposed into anatomical subgraphs for each streamline. In each subgraph, heterogeneous information from metapath-linked nodes is aggregated to obtain the final fiber representation. We evaluate the effectiveness of our framework on the HCP105 and TractoInferno datasets. The experimental results demonstrate that our method significantly outperforms previous state-of-the-art methods. The source code is available at https://github.com/CUHK-AIM-Group/Fiber-HGNN.
- 1 July 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 45 (7) , 3665-3676
Fiber HGNN: Heterogeneous Graph Neural Network for Fiber Tract Segmentation
Fiber tract segmentation is crucial for clinical applications such as brain function interpretation and surgical planning. Existing methods typically adopt either a cortical-parcellation-based or fiber clustering approach, but fail to simultaneously integrate heterogeneous information (e.g., streamline shape, point position, anatomical priors). In this work, we propose Fiber HGNN, a novel heterogeneous graph neural network that explicitly models and integrates heterogeneous information of fibers for accurate fiber tract segmentation. We construct a heterogeneous graph comprising three types of nodes: streamline, fiber keypoint and anatomical region. Specifically, fiber keypoints are representative points sampled along each streamline to characterize local geometric features, while anatomical regions provide contextual priors derived from brain atlas. This design enables the network to jointly capture the complementary information of streamline shape, local geometry, and anatomical priors, thus facilitating the learning of more discriminative feature representations. To further leverage implicit anatomical connectivity, we design a Metapath-guided Heterogeneous Information Aggregation (MHIA) network. By analyzing the spatial relationships between streamline keypoints and anatomical regions, the heterogeneous graph is decomposed into anatomical subgraphs for each streamline. In each subgraph, heterogeneous information from metapath-linked nodes is aggregated to obtain the final fiber representation. We evaluate the effectiveness of our framework on the HCP105 and TractoInferno datasets. The experimental results demonstrate that our method significantly outperforms previous state-of-the-art methods. The source code is available at https://github.com/CUHK-AIM-Group/Fiber-HGNN.
- 27 September 2025
- Published by Elsevier in Medical Image Analysis
- Vol. 107 (Pt A) , 103820
GAGM: Geometry-aware graph matching framework for weakly supervised gyral hinge correspondence
- 27 September 2025
- Published by Elsevier in Medical Image Analysis
- Vol. 107 (Pt A) , 103820
GAGM: Geometry-aware graph matching framework for weakly supervised gyral hinge correspondence
- 4 February 2025
- Published by Springer Nature in Brain Structure and Function
- Vol. 230 (2) , 1-14
Gyral peak variations between HCP and CHCP: functional and structural implications
Significant culture and ethnic diversity play an important role in shaping brain structure and function. Many attempts have been undertaken to connect ethnic variations with brain function, which, however, fluctuates over time and is costly, limiting its utility to identify consistent brain markers as well as its application to a broad population. In contrast, brain anatomy is less altered during a short period of time, but it is not fully understood whether it could serve as the ethnicity-sensitive landmark, or its variation is associated with functional one. In this study, We utilized gyral peaks, a set of early cortical folds, as cortical landmarks to explore the role of ethnic factors in brain anatomy and their relationship to brain function. Comparative experiments were conducted using the Human Connectome Project and the Chinese Human Connectome Project. In populations with similar ethnic backgrounds, gyral peak patterns showed greater consistency. For groups with significantly different ethnic backgrounds, we identified both shared peaks and peaks unique to each group. Compared to shared peaks, unique peaks showed significant differences in anatomical and functional network attributes and were spatially associated with working memory networks, which exhibited increased activation in their presence. Gene enrichment analysis provided additional support, suggesting that the unique peaks are associated with genes linked to working memory functions. These findings could provide new knowledge to understanding how ethnic diversity interplay with brain functions and associate with brain shapes.
- 4 February 2025
- Published by Springer Nature in Brain Structure and Function
- Vol. 230 (2) , 1-14
Gyral peak variations between HCP and CHCP: functional and structural implications
Significant culture and ethnic diversity play an important role in shaping brain structure and function. Many attempts have been undertaken to connect ethnic variations with brain function, which, however, fluctuates over time and is costly, limiting its utility to identify consistent brain markers as well as its application to a broad population. In contrast, brain anatomy is less altered during a short period of time, but it is not fully understood whether it could serve as the ethnicity-sensitive landmark, or its variation is associated with functional one. In this study, We utilized gyral peaks, a set of early cortical folds, as cortical landmarks to explore the role of ethnic factors in brain anatomy and their relationship to brain function. Comparative experiments were conducted using the Human Connectome Project and the Chinese Human Connectome Project. In populations with similar ethnic backgrounds, gyral peak patterns showed greater consistency. For groups with significantly different ethnic backgrounds, we identified both shared peaks and peaks unique to each group. Compared to shared peaks, unique peaks showed significant differences in anatomical and functional network attributes and were spatially associated with working memory networks, which exhibited increased activation in their presence. Gene enrichment analysis provided additional support, suggesting that the unique peaks are associated with genes linked to working memory functions. These findings could provide new knowledge to understanding how ethnic diversity interplay with brain functions and associate with brain shapes.
- 27 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 44 (10) , 4010-4022
FM-APP: Foundation Model for Any Phenotype Prediction via fMRI to sMRI Knowledge Transfer
Predicting individual-level non-neuroimaging phenotypes (e.g., fluid intelligence) using brain imaging data is a fundamental goal of neuroscience. Recent research has focused on utilizing high-cost functional magnetic resonance imaging (fMRI) to predict phenotypes seen during training. However, these methods 1) only consider predicting seen phenotypes, failing to achieve zero-shot inference for unseen phenotypes; 2) overlook the knowledge transfer from fMRI to structural MRI (sMRI), missing out on utilizing cost-effective sMRI for accurate predictions. To address these challenges, we propose a Foundational Model for Any Phenotype Prediction via fMRI to sMRI knowledge transfer (FM-APP), consisting of a Phenotypes Text Memory Bank (PTMB) module, Any Phenotype Prediction (APP) module, and fMRI to sMRI Knowledge Transfer (F2SKT) module. Our proposed FM-APP adapts to downstream tasks by generating regressor parameters instead of fine-tuning the model itself. Specifically, to retain important clues from seen phenotype descriptions, PTMB utilizes the BiomedCLIP model to store semantic features of seen phenotypes. To achieve any phenotype prediction, the APP introduces a regressor synthesizer for zero-shot inference. Additionally, to improve sMRI prediction accuracy while preserving its cost advantage, the F2SKT uses the PTMB to construct phenotype active maps, guiding adaptive knowledge transfer from fMRI to sMRI. Experiments on the Human Connectome Project (HCP) and HCP Aging datasets demonstrate our approach outperforms state-of-the-art methods, showcasing strong zero-shot inference capabilities and providing a novel framework for analyzing brain structure and phenotypes. Our code: https://github.com/ZhibinHe/FM-APP.
- 27 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 44 (10) , 4010-4022
FM-APP: Foundation Model for Any Phenotype Prediction via fMRI to sMRI Knowledge Transfer
Predicting individual-level non-neuroimaging phenotypes (e.g., fluid intelligence) using brain imaging data is a fundamental goal of neuroscience. Recent research has focused on utilizing high-cost functional magnetic resonance imaging (fMRI) to predict phenotypes seen during training. However, these methods 1) only consider predicting seen phenotypes, failing to achieve zero-shot inference for unseen phenotypes; 2) overlook the knowledge transfer from fMRI to structural MRI (sMRI), missing out on utilizing cost-effective sMRI for accurate predictions. To address these challenges, we propose a Foundational Model for Any Phenotype Prediction via fMRI to sMRI knowledge transfer (FM-APP), consisting of a Phenotypes Text Memory Bank (PTMB) module, Any Phenotype Prediction (APP) module, and fMRI to sMRI Knowledge Transfer (F2SKT) module. Our proposed FM-APP adapts to downstream tasks by generating regressor parameters instead of fine-tuning the model itself. Specifically, to retain important clues from seen phenotype descriptions, PTMB utilizes the BiomedCLIP model to store semantic features of seen phenotypes. To achieve any phenotype prediction, the APP introduces a regressor synthesizer for zero-shot inference. Additionally, to improve sMRI prediction accuracy while preserving its cost advantage, the F2SKT uses the PTMB to construct phenotype active maps, guiding adaptive knowledge transfer from fMRI to sMRI. Experiments on the Human Connectome Project (HCP) and HCP Aging datasets demonstrate our approach outperforms state-of-the-art methods, showcasing strong zero-shot inference capabilities and providing a novel framework for analyzing brain structure and phenotypes. Our code: https://github.com/ZhibinHe/FM-APP.
- 23 October 2024
- Published by Springer Nature
- p. 564-574
fTSPL: Enhancing Brain Analysis with FMRI-Text Synergistic Prompt Learning
- 23 October 2024
- Published by Springer Nature
- p. 564-574
fTSPL: Enhancing Brain Analysis with FMRI-Text Synergistic Prompt Learning
- 23 October 2024
- Published by Springer Nature
- p. 142-152
GBT: Geometric-Oriented Brain Transformer for Autism Diagnosis
- 23 October 2024
- Published by Springer Nature
- p. 142-152
GBT: Geometric-Oriented Brain Transformer for Autism Diagnosis
- 4 October 2024
- Published by Springer Nature
- p. 306-316
Hierarchical Graph Learning with Small-World Brain Connectomes for Cognitive Prediction
- 4 October 2024
- Published by Springer Nature
- p. 306-316
Hierarchical Graph Learning with Small-World Brain Connectomes for Cognitive Prediction
- Open Access
- 18 April 2024
- Published by eLife Sciences Publications, Ltd in eLife
- Vol. 12
Species -shared and -unique gyral peaks on human and macaque brains
Cortical folding is an important feature of primate brains that plays a crucial role in various cognitive and behavioral processes. Extensive research has revealed both similarities and differences in folding morphology and brain function among primates including macaque and human. The folding morphology is the basis of brain function, making cross-species studies on folding morphology important for understanding brain function and species evolution. However, prior studies on cross-species folding morphology mainly focused on partial regions of the cortex instead of the entire brain. Previously, our research defined a whole-brain landmark based on folding morphology: the gyral peak. It was found to exist stably across individuals and ages in both human and macaque brains. Shared and unique gyral peaks in human and macaque are identified in this study, and their similarities and differences in spatial distribution, anatomical morphology, and functional connectivity were also dicussed.
- 18 April 2024
- Published by eLife Sciences Publications, Ltd in eLife
- Vol. 12
Open AccessSpecies -shared and -unique gyral peaks on human and macaque brains
Cortical folding is an important feature of primate brains that plays a crucial role in various cognitive and behavioral processes. Extensive research has revealed both similarities and differences in folding morphology and brain function among primates including macaque and human. The folding morphology is the basis of brain function, making cross-species studies on folding morphology important for understanding brain function and species evolution. However, prior studies on cross-species folding morphology mainly focused on partial regions of the cortex instead of the entire brain. Previously, our research defined a whole-brain landmark based on folding morphology: the gyral peak. It was found to exist stably across individuals and ages in both human and macaque brains. Shared and unique gyral peaks in human and macaque are identified in this study, and their similarities and differences in spatial distribution, anatomical morphology, and functional connectivity were also dicussed.
- 29 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 28 (4) , 2223-2234
Brain Structural Connectivity Guided Vision Transformers for Identification of Functional Connectivity Characteristics in Preterm Neonates
Preterm birth is the leading cause of death in children under five years old, and is associated with a wide sequence of complications in both short and long term. In view of rapid neurodevelopment during the neonatal period, preterm neonates may exhibit considerable functional alterations compared to term ones. However, the identified functional alterations in previous studies merely achieve moderate classification performance, while more accurate functional characteristics with satisfying discrimination ability for better diagnosis and therapeutic treatment is underexplored. To address this problem, we propose a novel brain structural connectivity (SC) guided Vision Transformer (SCG-ViT) to identify functional connectivity (FC) differences among three neonatal groups: preterm, preterm with early postnatal experience, and term. Particularly, inspired by the neuroscience-derived information, a novel patch token of SC/FC matrix is defined, and the SC matrix is then adopted as an effective mask into the ViT model to screen out input FC patch embeddings with weaker SC, and to focus on stronger ones for better classification and identification of FC differences among the three groups. The experimental results on multi-modal MRI data of 437 neonatal brains from publicly released Developing Human Connectome Project (dHCP) demonstrate that SCG-ViT achieves superior classification ability compared to baseline models, and successfully identifies holistically different FC patterns among the three groups. Moreover, these different FCs are significantly correlated with the differential gene expressions of the three groups. In summary, SCG-ViT provides a powerfully brain-guided pipeline of adopting large-scale and data-intensive deep learning models for medical imaging-based diagnosis.
- 29 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 28 (4) , 2223-2234
Brain Structural Connectivity Guided Vision Transformers for Identification of Functional Connectivity Characteristics in Preterm Neonates
Preterm birth is the leading cause of death in children under five years old, and is associated with a wide sequence of complications in both short and long term. In view of rapid neurodevelopment during the neonatal period, preterm neonates may exhibit considerable functional alterations compared to term ones. However, the identified functional alterations in previous studies merely achieve moderate classification performance, while more accurate functional characteristics with satisfying discrimination ability for better diagnosis and therapeutic treatment is underexplored. To address this problem, we propose a novel brain structural connectivity (SC) guided Vision Transformer (SCG-ViT) to identify functional connectivity (FC) differences among three neonatal groups: preterm, preterm with early postnatal experience, and term. Particularly, inspired by the neuroscience-derived information, a novel patch token of SC/FC matrix is defined, and the SC matrix is then adopted as an effective mask into the ViT model to screen out input FC patch embeddings with weaker SC, and to focus on stronger ones for better classification and identification of FC differences among the three groups. The experimental results on multi-modal MRI data of 437 neonatal brains from publicly released Developing Human Connectome Project (dHCP) demonstrate that SCG-ViT achieves superior classification ability compared to baseline models, and successfully identifies holistically different FC patterns among the three groups. Moreover, these different FCs are significantly correlated with the differential gene expressions of the three groups. In summary, SCG-ViT provides a powerfully brain-guided pipeline of adopting large-scale and data-intensive deep learning models for medical imaging-based diagnosis.
- 1 October 2023
- Published by Springer Nature
- p. 548-558
$\mathrm {H^{2}}$GM: A Hierarchical Hypergraph Matching Framework for Brain Landmark Alignment- 1 October 2023
- Published by Springer Nature
- p. 548-558
$\mathrm {H^{2}}$GM: A Hierarchical Hypergraph Matching Framework for Brain Landmark Alignment