Decreased degree centrality values as a potential neuroimaging biomarker for migraine: A resting-state functional magnetic resonance imaging study and support vector machine analysis

Qian Wang, Yujun Gao, Yuandong Zhang, Xi Wang, Xuying Li, Hang Lin, Ling Xiong, Chunyan HuangFrontiers in Neurology 2023 · DOI

Abstract

Our findings demonstrate abnormal degree centrality (DC) values in the bilateral inferior temporal gyrus (ITG) among patients with migraine. These results provide insights into the neural mechanisms of migraines and suggest that abnormal DC values can serve as a potential neuroimaging biomarker for the diagnosis of migraine.

Introduction

Migraine is a prevalent and debilitating neurological disorder, yet its diagnosis remains largely subjective, relying on patient-reported symptoms. The lack of objective biomarkers—such as blood tests or standard structural MRI findings—often leads to misdiagnosis. While migraine is known to involve complex neural dysfunction, the specific imaging-based pathological mechanisms are poorly understood. This study aims to identify reliable "fingerprints" of the disease by analyzing the functional connectome—the network of communication between different brain regions.

Methods

The researchers utilized resting-state functional MRI (rs-fMRI) to capture spontaneous brain activity in 28 migraine patients and 27 healthy controls. They focused on Degree Centrality (DC), a graph-theory metric that measures the number of significant functional connections a specific brain region (node) has with the rest of the brain. High DC indicates a central "hub," while low DC suggests functional isolation.

The study then employed a Support Vector Machine (SVM), a machine learning classifier, to determine if these DC values could accurately distinguish patients from controls. Clinical severity was assessed using the Migraine Disability Assessment (MIDAS) and the Headache Impact Test-6 (HIT-6).

Key Insights and Results

The most significant finding was a marked decrease in DC within the bilateral inferior temporal gyrus (ITG) in migraine patients compared to healthy controls. This suggests that these regions, typically associated with visual and language processing, are becoming functionally isolated from the global brain network.

Figure 1: MRI slices showing decreased DC in the bilateral ITG and correlation with clinical disability scores.

Crucially, the researchers found a positive linear correlation between DC values in the left ITG and MIDAS scores. This indicates that the degree of network isolation in the brain directly reflects the level of disability experienced by the patient.

Using the DC values of the left ITG as a feature, the SVM classifier achieved a diagnostic accuracy of 81.82%, with a sensitivity of 85.71% and a specificity of 77.78%. These differences remained robust after Gaussian Random Field (GRF) correction ($P < 0.01$).

Discussion and Limitations

While these results are promising, several limitations must be addressed. The study's sample size (n=55) is relatively small, requiring validation in larger, multi-center trials. Additionally, most participants were taking medication, which could potentially influence brain connectivity patterns. Future research should also investigate whether these DC patterns differ across migraine subtypes (e.g., with vs. without aura) and determine the directionality of the relationship—whether decreased connectivity causes migraines or results from chronic attacks.

Despite these hurdles, the identification of decreased ITG centrality offers a potential path toward an objective, imaging-based diagnostic tool for migraine.

Summary updated

Abstract

Our findings demonstrate abnormal DC values in the bilateral ITG among patients with migraine, and the present results provide insights into the neural mechanism of migraines. The abnormal DC values can be used as a potential neuroimaging biomarker for the diagnosis of migraine.
BibTeX
@inproceedings{wang2023decreased,
  author = {Qian Wang and Yujun Gao and Yuandong Zhang and Xi Wang and Xuying Li and Hang Lin and Ling Xiong and Chunyan Huang},
  title = {{Decreased degree centrality values as a potential neuroimaging biomarker for migraine: A resting-state functional magnetic resonance imaging study and support vector machine analysis}},
  booktitle = {Frontiers in Neurology},
  year = {2023},
  doi = {10.3389/fneur.2022.1105592},
}
Figure