Scholarlink

Topological inference based on heat kernel estimation, persistent homology, and permutation testing has shown promise in tackling various modeling challenges associated with electroencephalography (EEG) from individuals with brain network disorders. In this paper, we propose a new heat kernel estimation/smoothing method of EEG signals through Chebyshev polynomials and a fast topological permutation test to compare persistent features of two groups of smoothed signals extracted through persistent homology. We also investigate the potential of the topological inference framework in a seizure lateralization problem.

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

Topological Inference for Seizure Lateralization

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Jian Yin,
  • Duc Anh Doan,
  • Sofia Kollia,
  • Andrew I. Yang,
  • Pavan Turaga,
  • Yuan Wang

摘要

Topological inference based on heat kernel estimation, persistent homology, and permutation testing has shown promise in tackling various modeling challenges associated with electroencephalography (EEG) from individuals with brain network disorders. In this paper, we propose a new heat kernel estimation/smoothing method of EEG signals through Chebyshev polynomials and a fast topological permutation test to compare persistent features of two groups of smoothed signals extracted through persistent homology. We also investigate the potential of the topological inference framework in a seizure lateralization problem.