Guest Editors
Min Zuo
Beijing Wuzi University; Beijing Technology and Business University
Dr. Min Zuo received his PhD in computer applications from the University of Science and Technology Beijing. He is currently a professor and master’s supervisor at Beijing Wuzi University and is also affiliated with the National Engineering Research Center for Agri-Product Quality Traceability at Beijing Technology and Business University. His earlier research focused on intelligent management systems and food quality evaluation using artificial intelligence techniques. More recently, his work has centered on deep learning approaches in cognitive neuroscience, particularly the use of multimodal data, such as fMRI and EEG, to investigate brain–behavior relationships under naturalistic paradigms. His research integrates neural network models with cognitive theories to decode perceptual, affective, and decision-making processes in real-world contexts.
Qingchuan Zhang
Beijing Technology and Business University
Dr. Qingchuan Zhang received his BS and MS degrees from Hebei University of Science and Technology and his PhD in computer science and technology from the University of Science and Technology Beijing. He is currently an associate professor in the School of Computer and Artificial Intelligence at Beijing Technology and Business University. His earlier research focused on semantic computing, natural language processing, and data mining, with applications in food quality monitoring and sensory evaluation. More recently, his work has centered on deep learning methodologies for cognitive neuroscience, with the goal of modeling brain–behavior relationships using naturalistic and multimodal datasets, including neuroimaging and linguistic response data. His current research contributes to the development of interpretable and scalable artificial intelligence frameworks for understanding higher-order cognitive functions and their neural substrates.
Wenjing Yan
Beijing Technology and Business University
Wenjing Yan received a B.E. degree in Electrical Engineering and Automation from Xi'an University of Technology in 2006 and a M.E. degree in Test and Measurement Technology and Instrumentation from China University of Mining and Technology (Beijing) in 2010. In 2017, she received a PhD in electrical and computer engineering from Auburn University. Nowadays, she is an associate professor at the National Engineering Research Center for Agricultural Product Quality Traceability at Beijing Technology and Business University. Her research interests include bioinformatics intelligent processing, brain signal processing, brain science research, image processing, deep learning, and pattern recognition. Her current work focuses on developing explainable deep learning models to uncover brain-behavior relationships from multimodal neuroimaging data.
Mingwen Bi
Beijing Wuzi University; Beijing Technology and Business University
Dr. Mingwen Bi received his PhD in computer science and technology from China Agricultural University. He is currently a lecturer in the Department of E-Commerce at Beijing Wuzi University. He has published five SCI-indexed papers as the first author, demonstrating his capability in conducting independent and impactful research. His research interests primarily focus on natural language processing and generative artificial intelligence, with a growing emphasis on cognitive neuroscience.Specifically, he explores how large language models and generative AI techniques can be leveraged to model brain signals and uncover brain–behavior relationships from naturalistic and multimodal data. He is also interested in using generative models for neural decoding and synthetic data generation to advance reproducible neuroscience. In addition, he maintains a secondary interest in smart logistics and supply chain management.Dr. Bi’s work bridges computational linguistics, generative AI, and cognitive science, contributing to interpretable and scalable models of human brain function.
Collection Overview
Multimodal neuroimaging (e.g., simultaneous EEG-fMRI, MRI-PET) has become a powerful approach to capturing complementary aspects of brain structure and function, offering high spatiotemporal resolution and rich biological information. However, the complexity and heterogeneity of multimodal data present significant challenges for conventional analysis methods. Deep learning has achieved remarkable success in multimodal fusion tasks, including cross-modal generation (EEG-to-fMRI synthesis), graph-based subtyping of Parkinson's disease and dementia, as well as diagnostic frameworks for Alzheimer's and Parkinson's diseases. Despite these advances, the "black-box" nature of deep neural networks severely limits their interpretability, posing a major barrier to clinical translation and fundamental neuroscientific understanding. Recent efforts in explainable AI (XAI) have begun to address this gap by developing attribution methods, attention mechanisms, and transparent architectures that reveal how models leverage specific modalities and brain regions to make predictions.
This Methods Collection aims to bring together reproducible protocols for multimodal neuroimaging using explainable deep learning, encompassing, but not limited to:
- XAI techniques for EEG-fMRI/MEG data fusion
- Interpretable graph neural networks for connectomic analysis
- Explainable AI methods for disease diagnosis
- Generative AI for neural decoding
- Multimodal brain imaging studies of hierarchical processing mechanisms, including spatiotemporal dynamics modeling, explainable perceptual representations, and cross-modal integration
- Real-time multimodal explainable analytics for clinically trustworthy decision making
- Multimodal neuroimaging-based approaches for brain connectome analysis
- Cross-scale brain dynamic system modeling
Articles
Abstracts
Tri-stream Disentangled Network for EEG-based Image Reconstruction via Adaptive Dual-space Fusion
Yipeng Zhou1,
Jingyuan Li1,
Kaizhong Zhao1,
Hong Zhu1,
Lin Zhu1,
Huan Jiang*1,
Qingchuan Zhang*1
1Beijing Technology and Business University
Multi-Dimensional Feature Adaptive Gating Network for Multi-View and Whole-Brain MRI-Based Alzheimer’s Disease Classification
Yipeng Zhou1,
Lin Zhu1,
Zhenqiao Liu1,
Qingchuan Zhang*1,
Jianli Wang1
1Beijing Technology and Business University
A Reproducible Neuro-Symbolic Protocol for Open-Vocabulary EEG-to-Text Decoding via Concept Grounding
Qingchuan Zhang1,
Kaizhong Zhao1,
Jingyuan Li1,
Hong Zhu1,
Jianli Wang*1,
Mingwen Bi*2
1Beijing Technology and Business University,
2Beijing Wuzi University
Functional Visual-Cortical Staging and Conservative Condition Fusion for Single-Trial fMRI Image Reconstruction
Hongbo Zhang*1,
Mingwen Bi*1,
Kailiang Wang*2,
Qingchuan Zhang*3
1Beijing Wuzi University,
2China National Center for Neurological Disorders,
3Beijing Technology and Business University
Revisiting Privacy Preservation in Brain-Computer Interfaces: Conceptual Boundaries, Risk Pathways, and a Protection-Strength Grading Framework
Lei Sun1,
Xiuqing Mao1,
Shuai Zhang1,
Qingyu Zeng*1,
Min Zhao1,
Jiyuan Li1,
Wenle Dong1
1PLA Information Engineering University