Scholarlink

<p>The brainstem is a highly conserved integrative hub that is essential for arousal, emotion, and motor control, which are functions profoundly affected in Parkinson’s disease (PD). However, low signal-to-noise ratios in neuroimaging of subcortical nuclei hinder the construction of brainstem parcellations that integrate complementary functional and structural information, limiting our understanding of circuit-specific pathology. Here, we developed the Multimodal Consensus Brainstem Parcellation (MCBP), a fine-grained atlas that integrates high-resolution naturalistic 7 T functional MRI and diffusion MRI acquired in the Human Connectome Project (<i>N</i> = 160). The MCBP exhibits high reproducibility, symmetry, and homogeneity. Using this parcellation, we find a pronounced structure–function dissociation within the brainstem. Furthermore, MCBP reveals a hierarchical brainstem organization aligned with the unimodal-to-transmodal axis, distinguishing motor execution from visually guided motor preparation that closely resembles the Somato-Cognitive Action Network (SCAN). Crucially, in an independent PD cohort, MCBP-derived connectivity improved disease classification and symptom-severity prediction compared with existing brainstem atlases, supporting its sensitivity to both categorical disease status and dimensional clinical variation. Feature decoding revealed a dual-system pattern of PD-related alterations across sensorimotor and cognitive–affective networks. Notably, it identified a specific brainstem–limbic circuit linked to emotional heterogeneity, providing circuit-level insights for distinguishing anxiety- and depression-related neuropsychiatric PD subtypes independent of motor severity. Beyond distinguishing and grading disease, the MCBP-derived connectivity measure also outperformed those derived from existing atlases in predicting multidimensional behavioral traits in a large healthy participant cohort (<i>N</i> = 976). Together, MCBP offers a fine-grained, biologically grounded brainstem parcellation and a framework for probing circuit-specific changes in neurodegenerative disorders.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A multimodal consensus parcellation of the human brainstem and its applications in Parkinson’s disease

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月30日

查看全文
  • Kexin Wang,
  • Tiantian Liu,
  • Yunxiao Ma,
  • Ziteng Han,
  • Xinyu Wu,
  • Xiu Wang,
  • Kai Zhang,
  • Guoyuan Yang,
  • Tianyi Yan

摘要

The brainstem is a highly conserved integrative hub that is essential for arousal, emotion, and motor control, which are functions profoundly affected in Parkinson’s disease (PD). However, low signal-to-noise ratios in neuroimaging of subcortical nuclei hinder the construction of brainstem parcellations that integrate complementary functional and structural information, limiting our understanding of circuit-specific pathology. Here, we developed the Multimodal Consensus Brainstem Parcellation (MCBP), a fine-grained atlas that integrates high-resolution naturalistic 7 T functional MRI and diffusion MRI acquired in the Human Connectome Project (N = 160). The MCBP exhibits high reproducibility, symmetry, and homogeneity. Using this parcellation, we find a pronounced structure–function dissociation within the brainstem. Furthermore, MCBP reveals a hierarchical brainstem organization aligned with the unimodal-to-transmodal axis, distinguishing motor execution from visually guided motor preparation that closely resembles the Somato-Cognitive Action Network (SCAN). Crucially, in an independent PD cohort, MCBP-derived connectivity improved disease classification and symptom-severity prediction compared with existing brainstem atlases, supporting its sensitivity to both categorical disease status and dimensional clinical variation. Feature decoding revealed a dual-system pattern of PD-related alterations across sensorimotor and cognitive–affective networks. Notably, it identified a specific brainstem–limbic circuit linked to emotional heterogeneity, providing circuit-level insights for distinguishing anxiety- and depression-related neuropsychiatric PD subtypes independent of motor severity. Beyond distinguishing and grading disease, the MCBP-derived connectivity measure also outperformed those derived from existing atlases in predicting multidimensional behavioral traits in a large healthy participant cohort (N = 976). Together, MCBP offers a fine-grained, biologically grounded brainstem parcellation and a framework for probing circuit-specific changes in neurodegenerative disorders.