Partial Similarity Reveals Dynamics in Brainstem-Midbrain Networks during Trigeminal Nociception

TY - JOUR

T1 - Partial Similarity Reveals Dynamics in Brainstem-Midbrain Networks during Trigeminal Nociception

AU - May, Arne

AU - Schulte, Laura Helene

AU - Nolte, Guido

AU - Mehnert, Jan

PY - 2020/9/2

Y1 - 2020/9/2

N2 - Imaging studies help us understand the important role of brainstem and midbrain regions in human trigeminal pain processing without solving the question of how these regions actually interact. In the current study, we describe this connectivity and its dynamics during nociception with a novel analytical approach called Partial Similarity (PS). We developed PS specifically to estimate the communication between individual hubs of the network in contrast to the overall communication within that network. Partial Similarity works on trial-to-trial variance of neuronal activity acquired with functional magnetic resonance imaging. It discovers direct communication between two hubs considering the remainder of the network as confounds. A similar method to PS is Representational Similarity, which works with ordinary correlations and does not consider any external influence on the communication between two hubs. Particularly the combination of Representational Similarity and Partial Similarity analysis unravels brainstem dynamics involved in trigeminal pain using the spinal trigeminal nucleus (STN)-the first relay station of peripheral trigeminal input-as a seed region. The combination of both methods can be valuable tools in discovering the network dynamics in fMRI and an important instrument for future insight into the nature of various neurological diseases like primary headaches.

AB - Imaging studies help us understand the important role of brainstem and midbrain regions in human trigeminal pain processing without solving the question of how these regions actually interact. In the current study, we describe this connectivity and its dynamics during nociception with a novel analytical approach called Partial Similarity (PS). We developed PS specifically to estimate the communication between individual hubs of the network in contrast to the overall communication within that network. Partial Similarity works on trial-to-trial variance of neuronal activity acquired with functional magnetic resonance imaging. It discovers direct communication between two hubs considering the remainder of the network as confounds. A similar method to PS is Representational Similarity, which works with ordinary correlations and does not consider any external influence on the communication between two hubs. Particularly the combination of Representational Similarity and Partial Similarity analysis unravels brainstem dynamics involved in trigeminal pain using the spinal trigeminal nucleus (STN)-the first relay station of peripheral trigeminal input-as a seed region. The combination of both methods can be valuable tools in discovering the network dynamics in fMRI and an important instrument for future insight into the nature of various neurological diseases like primary headaches.

KW - brainstem

KW - correlation

KW - fMRI

KW - pain

KW - pain processing

KW - partial correlation

KW - partial similarity

KW - representational similarity

U2 - 10.3390/brainsci10090603

DO - 10.3390/brainsci10090603

M3 - SCORING: Journal article

VL - 10

JO - BRAIN SCI

JF - BRAIN SCI

SN - 2076-3425

IS - 9

M1 - 603

ER -