BLBing Liu

Bing Liu

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Bing Liu has a total of 382 co-authors. Here are the top co-authors by number of shared publications:

  • Meng Wang

    State Key Laboratory of Cognitive Neuroscience and Learning, Beijing Normal University, Beijing, China

    23 shared publications
  • Yuqing Sun

    State Key Laboratory of Cognitive Neuroscience and Learning, Beijing Normal University, Beijing, China

    20 shared publications
  • Tianzi Jiang

    University of Chinese Academy of Sciences

    17 shared publications
  • Ke Hu

    Brainnetome Center and National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China

    13 shared publications
  • Xiaohan Tian

    State Key Laboratory of Cognitive Neuroscience and Learning, Beijing Normal University, Beijing 100091, China

    13 shared publications
  • Ang Li

    State Key Laboratory of Cognitive Science and Mental Health, Institute of Biophysics, Chinese Academy of Sciences, Beijing 100101, China

    12 shared publications

Latest Publications

  • Open Access

    Genetic mechanisms underlying the associations between brain structure and mental health

    by Ke Hu, Xiaohan Tian, Jing Lou, Chaoyue Ding, Tingting Zhang, Hao Zhang, Xiaoguang Zhang, Bing Liu

  • Open Access

    A generalizable biopsychosocial risk score for stratifying disease vulnerability in healthy populations

    by Jin Chen, Congying Chu, Miguel Garcia-Argibay, Wei Li, Christos Christogiannis, Tianye Jia, Charlotte Walton, Sangma Xie, Tifei Yuan, Samuele Cortese, et al.

  • Open Access

    Calcium signaling pathway implicates a shared genetic basis between psychiatric and cardiovascular diseases

    by Yunqing Zhu, Guorui Zhao, Yuyanan Zhang, Zhe Lu, Zhewei Kang, Xiaoyang Feng, Yundan Liao, Junyuan Sun, Rui Yuan, Yang Yang, et al.

    Psychiatric and cardiovascular diseases (CVDs) are frequently comorbid and are interconnected through the brain-heart axis. However, the underlying shared genetic etiology remains unknown in East Asians. To address this critical gap, we conducted a genome-wide pairwise trait pleiotropy study by leveraging genome-wide association studies of three major psychiatric disorders (schizophrenia [SCZ], bipolar disorder [BIP], major depressive disorder [MDD]) and ten cardiovascular traits (including eight CVDs) in East Asians. We identified genetic overlaps across seven disease pairs, such as SCZ with coronary artery disease. Through this pairwise approach, six of a total of 18 pleiotropic loci demonstrated tissue-specific expression in brain and cardiovascular systems. In the cross-ancestry replication, nine of the pleiotropic loci were validated. Among the novel pleiotropic genes, TPCN1, CACNA2D2, CACNA1D , and ATP2B1 are involved in voltage-dependent calcium channel activity, regulation of calcium influx, enriched in calcium-related pathway. We validated association with calcium signal pathway in an independent cohort. Calcium pathway-specific polygenic risk score for SCZ was associated with prolonged corrected QT (QTc) interval, which remained robust among individuals free from QTc-affecting drugs. Given that calcium-channel blockers are commonly prescribed for heart and blood vessel conditions, we performed drug target analysis by integrating gene expression profiles from the brain and cardiovascular tissues. Our findings implicated that calcium-channel blockers and peripheral vasodilators elevated SCZ risk, diuretics reduced the risks of SCZ, BIP, and MDD. Our study reveals extensive shared genetic architectures underlying psychiatric and CVDs, which warrant prudence in the use of calcium channel blockers among patients with concurrent psychiatric and CVDs.

  • Open Access

    From Psychological Distress to Multi-System Morbidity: A Transdiagnostic Internalizing Factor Predicts Chronic Disease Onset

    by Ang Li, Yini He, Tian Gao, Xiya Liu, Jiahong Wang, Junxing Xian, Yingjie Peng, Fengqian Xiao, Yan Yan, Lei Wang, et al.

    Physical health and mental well-being are bidirectionally linked. While the impact of somatic health on mental well-being is extensively investigated, the reciprocal “mind-to-body” trajectory—how psychological distress contributes to multi-system physical disease—remains underexplored, particularly in general populations. Here, we systematically investigated the biological pathways linking internalizing psychopathology to chronic disease onset by comparing the relative contributions of central (brain structural) and peripheral (multi-system biomarker) mediators. Using 14-year longitudinal data from 158,435 UK Biobank participants, we derived a transdiagnostic internalizing factor that prospectively predicted the onset of 28 chronic diseases across six physiological systems, with the highest-risk individuals experiencing disease onset up to 3.1 years earlier. Comprehensive phenotyping across 363 biomarkers revealed that peripheral biological dysregulation—particularly in the musculoskeletal and cardiovascular systems—exhibited associations with internalizing symptoms comparable to or stronger than brain structural alterations. Mediation analyses indicated that brain structure primarily mediated risks for neurological and cardiometabolic conditions, whereas peripheral biomarkers served as broad mediating pathways for a wider range of somatic outcomes, involving inflammation, metabolic dysregulation, and cardiovascular dysfunction. These findings reframe internalizing psychopathology as a systemic biological state, suggesting that integrating peripheral biological monitoring with dimensional psychopathology assessment could enhance risk stratification for chronic multimorbidity.

  • Open Access

    Shared genetic architecture between the topology of brain white matter structural connectome and fluid intelligence

    by Xinyi Dong, Weijie Huang, Hao-Jie Chen, Yunhao Zhang, Bing Liu, Daoqiang Zhang, Zhanjun Zhang, Guolin Ma, Ni Shu

    White matter (WM) connections support efficient interregional communication and form the structural basis of human fluid intelligence. However, the shared genetic architecture between WM structural connectome and fluid intelligence remains largely unknown. In this study, we analyzed diffusion-weighted MRI data from 26,655 UK Biobank participants to construct individual WM connectome and performed genome-wide association analyses on global and regional network topology. We identified 41 single nucleotide polymorphisms (SNPs) significantly associated with global efficiency and 45 SNPs linked to nodal efficiency. Genetic correlations with fluid intelligence were observed for 128 brain regions, with 44 and 3 regions sharing SNPs within chromosomes 6q21 and 3p21.1, respectively. Mendelian randomization revealed causal effects from WM connectome to fluid intelligence, particularly in the orbital and superior frontal gyrus. Finally, integrating polygenic scores with network efficiency improved the prediction of individual fluid intelligence. These findings highlight the genetic basis linking WM connectome topology and fluid intelligence, providing new insights into the neurogenetic underpinnings of fluid intelligence.


  • Open Access

    Decoupling MCI-specific signatures from shared neurobiological substrates of cognitive aging via deep learning

    by Biying Peng, Lin Du, Mingxi Dang, Ting Li, Ziyun Li, Jiawen Liu, Yaojing Chen, Bing Liu, Zhanjun Zhang

    The specific neuroanatomy of mild cognitive impairment (MCI) is obscured by its clinical heterogeneity and confounding effects from normative variation. This problem is compounded by the inability of conventional neuroimaging methods to disentangle these overlapping influences. Leveraging data from the Beijing Aging Brain Rejuvenation Initiative (BABRI, n = 918) and the Alzheimer’s Disease Neuroimaging Initiative (ADNI, n = 1293), this study employed a conditional variational autoencoder (CVAE) to: (1) systematically distinguish between aging-related cognitive decline and MCI-specific cognitive impairments; (2) implicitly disentangle latent, unknown confounding effects to identify MCI-specific structural brain alterations; and (3) construct individualized scores for predicting the risk of conversion to Alzheimer’s disease (AD). The CVAE effectively extracted MCI-specific latent features from T1 structural MRI, significantly correlated with episodic memory, attention, and executive function impairments. Reconstructions revealed characteristic deformation in regions including the middle and medial temporal lobes, frontal lobe, limbic system, and cerebellum. The robustness of this structural-cognitive impairment association model established in BABRI dataset was validated in the ADNI dataset. Moreover, predictive modeling using these features achieved superior AD-conversion prediction (AUC = 0.83) versus whole-brain atrophy (AUC = 0.74; p < 0.001) or CSF biomarkers (AUC = 0.77; p < 0.001).This work establishes a novel paradigm for isolating MCI-specific brain alterations from physiological aging.


  • Open Access

    Physical-Digital and social-nonsocial extracurricular engagement: differential effects on brain development and psychological outcomes in children

    by Xiaohan Tian, Ruoxin Yang, Jing Lou, Yuqing Sun, Meng Wang, Qi Wang, Wenkun Lei, Ang Li, Zhanjun Zhang, Bing Liu

    Extracurricular activities shape children’s development, yet how the interaction between their modality (digital/physical) and social context (social/non-social) impact mental health and neurodevelopment remains unclear. This study examined these relationships and their bidirectionality, as well as mediation by brain structural changes. Using baseline and one-year follow-up data from 8230 children in the Adolescent Brain Cognitive Development Study, we analyzed self-reported screen time for six digital activities (three social, three non-social), caregiver-reported participation in 28 extracurricular activities (11 social, 17 non-social), mental health via the Child Behavior Checklist and Kiddie Schedule for Affective Disorders and Schizophrenia, and gray matter volume using T1-weighted MRI. Distinct interaction patterns between modality and social context emerged. Physical-social activities were linked to fewer psychiatric symptoms, particularly withdrawn/depressed behaviors, and increased frontoparietal gray matter. In contrast, digital non-social activities were associated with higher psychiatric risk, especially rule-breaking behaviors, and reduced temporal gray matter. Physical non-social and digital-social activities demonstrated mixed effects on developmental measures. Longitudinal analyses revealed bidirectional activity-mental health associations, with brain volume changes mediating 3.7–5.0% of these relationships. These findings highlight the interactive role of activity modality and social context in children’s development and provide neurobiological evidence informing guidelines for beneficial developmental outcomes.


  • An Efficient Contrastive Deep Learning Model for Identifying Schizophrenia-Specific Neuroanatomical Variations

    by Lin Du, Biying Peng, Yuqing Sun, Chaoyue Ding, Xiaohan Tian, Wenkun Lei, Qi Wang, Xinghui Zhao, Yingjie Peng, Jing Lou, et al.

    Background and Hypothesis: Schizophrenia (SZ) is a debilitating mental disorder characterized by heterogeneous clinical manifestations and widespread brain structural abnormalities. However, the relationship between individual neuroanatomical abnormalities and clinical symptoms remains inconsistent. We hypothesize that isolating SZ-specific neuroanatomical variations could deepen our understanding of its pathophysiology and yield more reliable biomarkers for its clinical heterogeneity. Study Design: To investigate this, we developed DECODE-SZ (Dual Encoder Contrastive Decoding for Schizophrenia), a novel model combining contrastive learning, 3D convolutional neural networks, and variational autoencoders (VAE) to isolate SZ-specific neuroanatomical features. We applied this model to structural MRI data from 641 patients diagnosed with SZ and 609 healthy controls across 8 independent sites in China, employing a leave-one-site-out cross-validation approach to ensure robust and unbiased results. Our analysis focused on examining the relationship between SZ-specific gray matter alterations and clinical symptoms (measured by PANSS scores), while also considering non-clinical variables such as age, sex, and education. Study Results: The DECODE-SZ model successfully extracted SZ-specific gray matter features, revealing that these features, rather than common variations with controls, were more strongly associated with PANSS scores. Consistent brain regions exhibiting these alterations were identified across multiple sites, supporting the reliability of the findings. A control experiment using a traditional VAE model demonstrated the superior performance of DECODE-SZ in isolating meaningful SZ-specific neuroanatomical variations. Conclusions: Our findings highlight the potential of SZ-specific neuroanatomical alterations as key biomarkers for clinical outcomes in SZ. DECODE-SZ offers a promising tool for advancing the understanding of SZ and may inform future diagnostic and therapeutic strategies.

  • Top Cited Papers
    Open Access

    Multimodal neuroimaging and AI integration in cognitive disorders: advances, challenges, and future directions for precision medicine

    by Mingxi Dang, Bing Liu, Yaojing Chen, Zhanjun Zhang

    Cognitive disorders, with dementia as a primary exemplar, present profound diagnostic and therapeutic challenges due to their complex pathologies and heterogeneous presentations. Artificial intelligence (AI), particularly when applied to multimodal neuroimaging and clinical data, offers a powerful approach to advancing precision medicine in this domain. This comprehensive review first examines foundational AI algorithms, including artificial neural networks for feature extraction, multimodal fusion strategies (e.g. early, intermediate, and late fusion) for data integration, and explainable AI (XAI) techniques to enhance clinical transparency. The core focus is on the application of these multimodal AI frameworks across the dementia care continuum, encompassing improved differential diagnosis, early detection through presymptomatic biomarkers, development of predictive models for disease progression, and optimization of patient stratification for clinical trials. Despite significant advances, persistent challenges remain, including limited generalizability across populations and protocols, data scarcity for non-Alzheimer’s dementias and prodromal stages—exacerbated by demographic biases—and barriers to interpretability. We discuss solutions such as federated learning for privacy-preserving data sharing and advanced XAI techniques. Finally, we outline pivotal future directions, including intelligent sensor fusion for discovering novel early biomarkers, hybrid AI architectures combining generative and discriminative models, innovations for handling missing modalities, and robust multicenter data integration frameworks. By synthesizing these advances, this review highlights the role of multimodal AI in advancing precise diagnosis, early prediction, and therapeutic development for neurodegenerative and vascular cognitive disorders, while identifying key translational challenges for precision medicine.


  • Open Access

    A Biopsychosocial Risk Score for Stratifying Disease Vulnerability in Healthy Populations: A Prospective Cohort and Multi-Omics Study in the UK Biobank

    by Jin Chen, Congying Chu, Miguel Garcia-Argibay, Wei Li, Christos Christogiannis, Tianye Jia, Charlotte Walton, Sangma Xie, Tifei Yuan, Samuele Cortese, et al.

    Proactive identification of systemic vulnerability for disease(s) before clinical onset in healthy individuals is an ultimate goal of preventive and precision medicine, yet current tools remain largely disease-specific and fail to quantify latent vulnerability, an integrative measure of underlying health status, for early prevention and risk-stratified intervention. To address this, we developed the Risk Score for Disease Vulnerability (RS4DV) based on 85 accessible biopsychosocial measures, which was constructed using a Light Gradient Boosting Machine trained and validated in the UK Biobank ( n = 391,193). Its capacity to capture pre-clinical vulnerability was subsequently evaluated in a held-out cohort free of baseline diagnoses ( n = 35,193). Over a median follow-up of 14.7 years, baseline RS4DV stratified long-term health outcomes in the held-out cohort, in which high-risk individuals exhibited accelerated disease accumulation (HR = 2.53, 95% CI: 2.44-2.62) and elevated risk of all-cause mortality (HR = 4.03, 95% CI: 3.68-4.41). Multi-omics analyses further revealed that RS4DV captures signatures of systemic inflammation, metabolic dysregulation, and accelerated brain ageing, establishing its biological interpretability. To facilitate real-world translation, we developed a practically feasible version of the RS4DV with only six routinely accessible items, which maintained robust predictive fidelity relative to the full model. This light-version model demonstrated robust generalizability in the 1970 British Birth Cohort under a zero-shot paradigm. Collectively, RS4DV provides a biologically grounded and scalable tool for personalized risk assessment decades before clinical onset and early-stage risk stratification, enabling a paradigm shift toward proactive health management and precision prevention.