Scholarlink

<p>Alzheimer’s disease is a neurodegenerative disorder characterized by impairments in memory, cognition, and behavior. Resting-state functional magnetic resonance imaging is one of the valuable techniques for studying brain function in Alzheimer’s disease, providing insights into neural mechanisms and aiding early diagnosis. Conventional analytical approaches for such data, including dynamic functional connectivity and time series analysis, often explore the spatiotemporal relationships between brain regions. However, the short duration of real-world functional magnetic resonance imaging data often limits their ability to capture nonlinear features that are critical for early-stage Alzheimer’s disease detection. Here we show that a surrogate model based on reservoir computing and compressed sensing can be used to expand data through time series predictions and extract nonlinear dynamical characteristics of Alzheimer’s disease. Two indicators are further proposed based on the developed model, the maximum Lyapunov exponent and the phase locking values, indicating a reduced dynamical complexity in Alzheimer’s disease and the key Alzheimer’s disease-affected brain regions in the frontal and parietal lobes, respectively. The classification accuracy based on these indicators can reach as high as 87% across different datasets, with a mean accuracy of 73.1% &#xa0;± 10.9%. This framework improves the use of functional magnetic resonance imaging data and may support further theoretical investigations in neurodegenerative disease research.</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

Characterizing Alzheimer’s disease with reservoir computing

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月15日

查看全文
  • Qing Li,
  • Zijian Wang,
  • Yihe Zhang,
  • Zixiang Yan,
  • Jian Gao,
  • Xingang Wang,
  • Yong Liu,
  • Jinghua Xiao

摘要

Alzheimer’s disease is a neurodegenerative disorder characterized by impairments in memory, cognition, and behavior. Resting-state functional magnetic resonance imaging is one of the valuable techniques for studying brain function in Alzheimer’s disease, providing insights into neural mechanisms and aiding early diagnosis. Conventional analytical approaches for such data, including dynamic functional connectivity and time series analysis, often explore the spatiotemporal relationships between brain regions. However, the short duration of real-world functional magnetic resonance imaging data often limits their ability to capture nonlinear features that are critical for early-stage Alzheimer’s disease detection. Here we show that a surrogate model based on reservoir computing and compressed sensing can be used to expand data through time series predictions and extract nonlinear dynamical characteristics of Alzheimer’s disease. Two indicators are further proposed based on the developed model, the maximum Lyapunov exponent and the phase locking values, indicating a reduced dynamical complexity in Alzheimer’s disease and the key Alzheimer’s disease-affected brain regions in the frontal and parietal lobes, respectively. The classification accuracy based on these indicators can reach as high as 87% across different datasets, with a mean accuracy of 73.1%  ± 10.9%. This framework improves the use of functional magnetic resonance imaging data and may support further theoretical investigations in neurodegenerative disease research.