Skip to main content

ORIGINAL RESEARCH article

Front. Neurosci.
Sec. Brain Imaging Methods
Volume 18 - 2024 | doi: 10.3389/fnins.2024.1488193

A three-classification model for identifying migraine with right-to-left shunt using lateralization of functional connectivity and brain network topology: a resting-state fMRI study

Provisionally accepted
Weifang Nie Weifang Nie 1*Weiming Zeng Weiming Zeng 1Jiajun Yang Jiajun Yang 2Lei Wang Lei Wang 1Yuhu Shi Yuhu Shi 1
  • 1 Shanghai Maritime University, pudong, China
  • 2 Shanghai Jiao Tong University Affiliated Sixth People’s Hospital, shanghai, China

The final, formatted version of the article will be published soon.

    Right-to-left shunting has been significantly associated with migraine, although the neural mechanisms remain complex and not fully elucidated. The aim of this study was to investigate the variability of brain asymmetry in individuals with migraine with right-to-left shunting, migraine without right-to-left shunting and normal controls using resting-state fMRI technology and to construct a three-classification model. Firstly, asymmetries in functional connectivity and brain network topology were quantified to laterality indices. Secondly, the laterality indices were employed to construct a three-classification model using decision tree and random forest algorithms. Ultimately, through a feature score analysis, the key brain regions that contributed significantly to the classification were extracted, and the associations between these brain regions and clinical features were investigated. Our experimental results showed that the initial classification accuracy reached 0.8961. Subsequently, validation using an independent sample set resulted in a classification accuracy of 0.8874. Further, after expanding the samples by the segmentation strategy, the classification accuracies were improved to 0.9103 and 0.9099. Additionally, the third sample set yielded a classification accuracy of 0.8745. Finally, 9 pivotal brain regions were identified and distributed in the default network, the control network, the visual network, the limbic network, the somatomotor network and the salience/ventral attention network. The results revealed distinct lateralization features in the brains of the three groups, which were closely linked to migraine and right-to-left shunting symptoms and could serve as potential imaging biomarkers for clinical diagnosis. Our findings enhanced our understanding of migraine and right-to-left shunting three-classification model for identifying migraine 2 mechanisms and offered insights into assisting clinical diagnosis.

    Keywords: Laterality Index, fMRI, right-to-left shunt, Migraine, Classification 1

    Received: 29 Aug 2024; Accepted: 28 Oct 2024.

    Copyright: © 2024 Nie, Zeng, Yang, Wang and Shi. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

    * Correspondence: Weifang Nie, Shanghai Maritime University, pudong, China

    Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.