Scholarlink

This study investigates the detection of Autism Spectrum Disorder (ASD) by analyzing brain connectivity patterns derived from resting-state fMRI data in the ABIDE dataset. Using the CC400 atlas and a Deep Neural Network (DNN), the model achieved an impressive accuracy of 100% in both cross-validation and on a test set comprising 30% of the dataset. The dataset included 403 individuals with ASD and 468 typically developing controls (TC). This performance surpasses current state-of-the-art methods, emphasizing the effectiveness of advanced feature selection and deep learning approaches in classifying ASD based on brain connectivity pattern.

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

Autism Spectrum Disorder Detection Through Connectivity Analysis and Feature Selection

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Imane Chkifa,
  • Moncef Garouani,
  • Aicha Majda

摘要

This study investigates the detection of Autism Spectrum Disorder (ASD) by analyzing brain connectivity patterns derived from resting-state fMRI data in the ABIDE dataset. Using the CC400 atlas and a Deep Neural Network (DNN), the model achieved an impressive accuracy of 100% in both cross-validation and on a test set comprising 30% of the dataset. The dataset included 403 individuals with ASD and 468 typically developing controls (TC). This performance surpasses current state-of-the-art methods, emphasizing the effectiveness of advanced feature selection and deep learning approaches in classifying ASD based on brain connectivity pattern.