TBTianyu Bai

Tianyu Bai

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Tianyu Bai has a total of 29 co-authors. Here are the top co-authors by number of shared publications:

  • Guozheng Feng

    State Key Laboratory of Cognitive Neuroscience and Learning & IDG/McGovern Institute for Brain Research, Beijing Normal University, Beijing, China

    3 shared publications
  • Hao-Jie Chen

    State Key Laboratory of Cognitive Neuroscience and Learning & IDG/McGovern Institute for Brain Research, Beijing Normal University, Beijing, China

    3 shared publications
  • Ni Shu

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

    3 shared publications
  • Yichen Wang

    State Key Laboratory of Cognitive Neuroscience and Learning & IDG/McGovern Institute for Brain Research, Beijing Normal University, Beijing, China; BABRI Centre, Beijing Normal University, Beijing, China; Beijing Key Laboratory of Brain Imaging and Connectomics, Beijing Normal University, Beijing, China

    3 shared publications
  • Mingkai Zhang

    Department of Neurology, Xuanwu Hospital of Capital Medical University, Beijing, China

    2 shared publications
  • Weijie Huang

    College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, the Key Laboratory of Brain-Machine Intelligence Technology, Ministry of Education, Nanjing, China.

    2 shared publications

Latest Publications

  • Open Access

    Multimodal integration of plasma biomarkers, MRI, and genetic risk to predict cerebral amyloid burden in Alzheimer’s disease

    by Yichen Wang, Hao-Jie Chen, Yuxin Cheng, YaoXin Xie, Yuyan Cheng, Shiyun Zhao, Yidong Jiang, Tianyu Bai, Yanxi Huo, Kexin Wang, et al.


  • Open Access

    Polygenic risk for Alzheimer’s disease in healthy aging: age-related and APOE-driven effects on brain structures and cognition

    by Hao-Jie Chen, Xinyi Dong, Yichen Wang, Kexin Wang, Guozheng Feng, Tianyu Bai, Mingkai Zhang, Kaiyu Gan, Jun-Jie Peng, Weijie Huang, et al.

    Alzheimer’s disease (AD) is characterized by progressive neurodegeneration and cognitive decline with age. The genetic architecture of AD involves multiple loci, including the apolipoprotein E gene (APOE). The polygenic risk scores for AD (AD-PRS) provide a comprehensive genome-wide assessment of AD risk, yet their age-related effects on brain structures and cognitive function in cognitively unimpaired individuals remain largely undefined. We analyzed cognitively unimpaired, genetically unrelated Caucasians from the UK Biobank (N = 21,236, 64.5 ± 7.6 years). AD-PRS was derived using a Bayesian approach incorporating approximately 5 million genetic variants (UK Biobank’s standard PRS). Brain structures were measured with regional gray matter (GM) volumes and tract-wise microstructural white matter (WM) integrity. Cognitive performance was evaluated with executive function, visuospatial function, reasoning, and memory. Sliding window analyses were performed to investigate age-related polygenic effects on brain structures, and mediation analyses tested whether structural changes mediated the gene–cognition relationship across different age groups. Analyses were replicated using two custom PRSs—one including APOE and the other excluding APOE regions—calculated with the clumping-and-thresholding approach. High AD-PRS was associated with accelerated GM atrophy (particularly in the hippocampus, thalamus, and parahippocampus), increased cerebral ventricular volume, and reduced WM integrity (especially in the fornix, cingulum, and superior fronto-occipital fasciculus). These polygenic effects demonstrated significant age-related amplification (pBonf < 0.05), with the strongest effects in individuals aged ≥ 75. Elevated AD-PRS was linked to lower cognitive performance across aging, especially in executive function, reasoning, and memory, which were significantly mediated by structural brain changes in subcortical and posterior limbic regions and their WM connections, predominantly in late aging (p < 0.05). Sensitivity analyses confirmed the robustness of these findings, emphasizing the dominant contribution of APOE, while also identifying age-specific effects from non-APOE variants. High polygenic risk for AD may be associated with accelerated cognitive decline in healthy aging, mediated by structural changes within hippocampal-thalamic regions and their connecting WM tracts. We provide insights into the early pathogenesis of AD and support the potential for age-targeted screening and early intervention for individuals at high genetic risk.


  • Open Access

    Multimodal Integration of Alzheimer's Plasma Biomarkers, MRI, and Genetic Risk for Individual Prediction of Cerebral Amyloid Burden

    by Wang Yichen, Hao-Jie Chen, Cheng Yuxin, Cheng Yuyan, Zhao Shiyun, Wang Kexin, Jiang Yidong, Bai Tianyu, Huo Yanxi, Zhang MingKai, et al.

    Alzheimer's disease (AD), the most prevalent neurodegenerative disorder, is marked by the accumulation of amyloid-β (Aβ) plaques. Although cerebral Aβ positron emission tomography (Aβ-PET) remains the gold standard for assessing cerebral Aβ burden, its clinical utility is hindered by cost, radiation exposure, and limited availability. Plasma biomarkers serve as promising non-invasive predictors of cerebral Aβ burden, but reliance on a single marker often leads to suboptimal predictive performance. To address this, we proposed a multimodal machine learning strategy that integrates readily accessible and non-invasive features—such as plasma biomarkers, structural magnetic resonance imaging (sMRI)-derived atrophy measures, diffusion tensor imaging (DTI)-based structural connectomes (SCs), and genetic risk profiles—to predict cerebral Aβ burden and evaluate the relative contribution of each modality to predictive performance. Specifically, a random forest regressor was trained using data from the Alzheimer's Disease Neuroimaging Initiative (ADNI; n = 150) and evaluated with leave-one-out cross-validation. Our results showed that integrating multimodal features improves the predictive power on cerebral amyloid burden: while the baseline model using plasma and clinical variables alone achieved an R2 of 0.52, adding neuroimaging and apolipoprotein E (APOE) genotype features improved performance (R2 = 0.617), and replacing APOE with polygenic risk scores (PRS) further enhanced accuracy (R2 = 0.637). The predictive value of multimodal integration was also replicated in an independent cohort (SILCODE; n = 101). Moreover, a multiclass classifier trained with the same multimodal features achieved high accuracy in distinguishing clinical stages of Aβ burden—normal controls (NC), mild cognitive impairment (MCI), and Alzheimer's disease (AD)—with area under the curve (AUC) values of 0.86, 0.77, and 0.93, respectively. These findings highlight the value of combining plasma, imaging, and genetic data to non-invasively estimate cerebral Aβ burden, offering a potential alternative to PET imaging for early AD risk assessment.