Whole‐brain functional gradients reveal cortical and subcortical alterations in patients with episodic migraine

Chae Hyeon Lee,Hyunjin Park,Mi Ji Lee,Bo‐yong Park

DOI: https://doi.org/10.1002/hbm.26204

IF: 4.8

2023-01-19

Human Brain Mapping

Abstract:We investigated functional connectome alterations in patients with migraine using dimensionality reduction techniques. We found significant between‐group differences in low‐dimensional representations of functional connectivity between patients with migraine and healthy controls, particularly in the sensory/motor and limbic cortices. Supervised machine learning indicated that imaging features of sensory/motor regions play an important role in predicting migraine frequency. Migraine is a type of headache with multiple neurological symptoms. Prior neuroimaging studies in patients with migraine based on functional magnetic resonance imaging have found regional as well as network‐level alterations in brain function. Here, we expand on prior studies by establishing whole‐brain functional connectivity patterns in patients with migraine using dimensionality reduction techniques. We studied functional brain connectivity in 50 patients with episodic migraine and sex‐ and age‐matched healthy controls. Using dimensionality reduction techniques that project high‐dimensional functional connectivity onto low‐dimensional representations (i.e., eigenvectors), we found significant between‐group differences in the eigenvectors between patients with migraine and healthy controls, particularly in the sensory/motor and limbic cortices. Furthermore, we assessed between‐group differences in subcortical connectivity with subcortical weighted manifolds defined by subcortico‐cortical connectivity multiplied by cortical eigenvectors and revealed significant alterations in the amygdala. Finally, leveraging supervised machine learning, we moderately predicted headache frequency using cortical and subcortical functional connectivity features, again indicating that sensory and limbic regions play a particularly important role in predicting migraine frequency. Our study confirmed that migraine is a hierarchical disease of the brain that shows alterations along the sensory‐limbic axis, and therefore, the functional connectivity in these areas could be a useful marker to investigate migraine symptomatology.

radiology, nuclear medicine & medical imaging,neurosciences,neuroimaging