Scholarlink

Background <p>Neurological conditions account for millions of deaths per year and induce long-lasting cognitive impairments. The disruption of structural brain networks predicts the emergence of cognitive impairments in stroke cases, but the role of the white matter in modeling longitudinal behavioral trajectories in glioma patients is understudied.</p> Methods <p>We analyzed 486 intracranial brain stimulations from 297 patients (age range 37–40, male ratio 53–64% depending on the functional categories) along with functional and structural brain connectivity data from over 1750 healthy individuals, to create a network mapping method able to identify the neural substrate causally involved in language production. We tested the validity of our procedure by (i) quantifying the spatial correspondence between white matter metabolic and hemodynamic spontaneous activity, measured via resting-state functional Magnetic Resonance Imaging and [18 F]-fluorodeoxyglucose functional Positron Emission Tomography (respectively); (ii) predicting unseen intracranial stimulations points; (iii) modeling the severity of stroke-induced aphasia (<i>n</i> = 105) and the longitudinal recovery of language abilities in glioma patients (<i>n</i> = 42, 3 timepoints).</p> Results <p>We show that spontaneous white matter hemodynamic oscillations map into metabolic fluctuations. We also demonstrate that the integration of patient-specific intracranial stimulation points and normative human connectivity data (i) is predictive of unseen stimulation points; (ii) provides better estimates than total lesion volume in predicting the severity of stroke-induced aphasia symptoms; (iii) models post-operative language recovery trajectories better than state-of-the-art clinical measures in glioma patients.</p> Conclusions <p>This work presents a data-driven and neurobiologically grounded tool for modeling cognitive and neurological impairments in terms of network disruption, demonstrating improved precision over existing approaches.</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

Integrating direct electrical stimulation with brain connectivity predicts lesion-induced language impairment and recovery

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年10月6日

查看全文
  • Ludovico Coletta,
  • Paolo Avesani,
  • Luca Zigiotto,
  • Martina Venturini,
  • Luciano Annicchiarico,
  • Laura Vavassori,
  • Sharna D. Jamadar,
  • Emma X. Liang,
  • Justine Y. Hansen,
  • Bratislav Misic,
  • Sam Ng,
  • Hugues Duffau,
  • Silvio Sarubbo

摘要

Background

Neurological conditions account for millions of deaths per year and induce long-lasting cognitive impairments. The disruption of structural brain networks predicts the emergence of cognitive impairments in stroke cases, but the role of the white matter in modeling longitudinal behavioral trajectories in glioma patients is understudied.

Methods

We analyzed 486 intracranial brain stimulations from 297 patients (age range 37–40, male ratio 53–64% depending on the functional categories) along with functional and structural brain connectivity data from over 1750 healthy individuals, to create a network mapping method able to identify the neural substrate causally involved in language production. We tested the validity of our procedure by (i) quantifying the spatial correspondence between white matter metabolic and hemodynamic spontaneous activity, measured via resting-state functional Magnetic Resonance Imaging and [18 F]-fluorodeoxyglucose functional Positron Emission Tomography (respectively); (ii) predicting unseen intracranial stimulations points; (iii) modeling the severity of stroke-induced aphasia (n = 105) and the longitudinal recovery of language abilities in glioma patients (n = 42, 3 timepoints).

Results

We show that spontaneous white matter hemodynamic oscillations map into metabolic fluctuations. We also demonstrate that the integration of patient-specific intracranial stimulation points and normative human connectivity data (i) is predictive of unseen stimulation points; (ii) provides better estimates than total lesion volume in predicting the severity of stroke-induced aphasia symptoms; (iii) models post-operative language recovery trajectories better than state-of-the-art clinical measures in glioma patients.

Conclusions

This work presents a data-driven and neurobiologically grounded tool for modeling cognitive and neurological impairments in terms of network disruption, demonstrating improved precision over existing approaches.