Comparison between a pure functional connectivity and a mixed functional-topological model in functional connectivity. An application on parahippocampal gyrus-anterior division data

P. Finotelli, M. Cabinio, Ottavia Dipasquale, Mara Cercignani, Baglio Francesca, P. Dulio

Research output: Contribution to journalArticle

Abstract

The investigation of brain functional connectivity (FC) by means of rsfMRI techniques is on-going challenge in the neuroimaging field. In the present investigation we compare two mathematical models for the computation of Resting-State Networks: The first one is based on the pure analysis of time courses (pure Functional Connectivity, pFC), the second one, the FD model, includes both the time courses, the anatomical information between brain nodes and topological metrics of the network which models the brain. The two approaches have been evaluated by comparing the maximal weights of the links representing the neural network obtained by applying the two models to a dataset of rsfMRI images from 133 healthy subjects. FC analyses were performed using the two methods by focusing on the functional connections between the anterior part of the parahippocampal gyrus (PHGA), a core area for cognitive and emotive processes, and the rest of the brain. Based on the literature, we expect to collect evidences of the role of PHGA as connector hub between the temporal pole and the nodes of the Default Model Network. As expected, the majority of the significant links involved were highlighted by both methods. However, the FD model highlighted a greater number of links compared with pFC. These links are in line with the known neuroanatomy. Hence, our results invite to consider the FD approach as an effective approach of analysis, since due to its characteristics it could provide a more complete description of the brain network.

Original languageEnglish
Article number101570
JournalBiomedical Signal Processing and Control
Volume54
DOIs
Publication statusPublished - Sep 1 2019

Fingerprint

Parahippocampal Gyrus
Brain
Neuroimaging
Neuroanatomy
Poles
Healthy Volunteers
Theoretical Models
Mathematical models
Neural networks
Weights and Measures

Keywords

  • Brain networks
  • Default mode network
  • Functional connectivity
  • Functional magnetic resonance imaging
  • Parahippocampal gyrus anterior division
  • Resting state
  • Temporal pole

ASJC Scopus subject areas

  • Signal Processing
  • Health Informatics

Cite this

@article{c6030b11a26d499f92a483284e6b928f,
title = "Comparison between a pure functional connectivity and a mixed functional-topological model in functional connectivity. An application on parahippocampal gyrus-anterior division data",
abstract = "The investigation of brain functional connectivity (FC) by means of rsfMRI techniques is on-going challenge in the neuroimaging field. In the present investigation we compare two mathematical models for the computation of Resting-State Networks: The first one is based on the pure analysis of time courses (pure Functional Connectivity, pFC), the second one, the FD model, includes both the time courses, the anatomical information between brain nodes and topological metrics of the network which models the brain. The two approaches have been evaluated by comparing the maximal weights of the links representing the neural network obtained by applying the two models to a dataset of rsfMRI images from 133 healthy subjects. FC analyses were performed using the two methods by focusing on the functional connections between the anterior part of the parahippocampal gyrus (PHGA), a core area for cognitive and emotive processes, and the rest of the brain. Based on the literature, we expect to collect evidences of the role of PHGA as connector hub between the temporal pole and the nodes of the Default Model Network. As expected, the majority of the significant links involved were highlighted by both methods. However, the FD model highlighted a greater number of links compared with pFC. These links are in line with the known neuroanatomy. Hence, our results invite to consider the FD approach as an effective approach of analysis, since due to its characteristics it could provide a more complete description of the brain network.",
keywords = "Brain networks, Default mode network, Functional connectivity, Functional magnetic resonance imaging, Parahippocampal gyrus anterior division, Resting state, Temporal pole",
author = "P. Finotelli and M. Cabinio and Ottavia Dipasquale and Mara Cercignani and Baglio Francesca and P. Dulio",
year = "2019",
month = "9",
day = "1",
doi = "10.1016/j.bspc.2019.101570",
language = "English",
volume = "54",
journal = "Biomedical Signal Processing and Control",
issn = "1746-8094",
publisher = "Elsevier BV",

}

TY - JOUR

T1 - Comparison between a pure functional connectivity and a mixed functional-topological model in functional connectivity. An application on parahippocampal gyrus-anterior division data

AU - Finotelli, P.

AU - Cabinio, M.

AU - Dipasquale, Ottavia

AU - Cercignani, Mara

AU - Francesca, Baglio

AU - Dulio, P.

PY - 2019/9/1

Y1 - 2019/9/1

N2 - The investigation of brain functional connectivity (FC) by means of rsfMRI techniques is on-going challenge in the neuroimaging field. In the present investigation we compare two mathematical models for the computation of Resting-State Networks: The first one is based on the pure analysis of time courses (pure Functional Connectivity, pFC), the second one, the FD model, includes both the time courses, the anatomical information between brain nodes and topological metrics of the network which models the brain. The two approaches have been evaluated by comparing the maximal weights of the links representing the neural network obtained by applying the two models to a dataset of rsfMRI images from 133 healthy subjects. FC analyses were performed using the two methods by focusing on the functional connections between the anterior part of the parahippocampal gyrus (PHGA), a core area for cognitive and emotive processes, and the rest of the brain. Based on the literature, we expect to collect evidences of the role of PHGA as connector hub between the temporal pole and the nodes of the Default Model Network. As expected, the majority of the significant links involved were highlighted by both methods. However, the FD model highlighted a greater number of links compared with pFC. These links are in line with the known neuroanatomy. Hence, our results invite to consider the FD approach as an effective approach of analysis, since due to its characteristics it could provide a more complete description of the brain network.

AB - The investigation of brain functional connectivity (FC) by means of rsfMRI techniques is on-going challenge in the neuroimaging field. In the present investigation we compare two mathematical models for the computation of Resting-State Networks: The first one is based on the pure analysis of time courses (pure Functional Connectivity, pFC), the second one, the FD model, includes both the time courses, the anatomical information between brain nodes and topological metrics of the network which models the brain. The two approaches have been evaluated by comparing the maximal weights of the links representing the neural network obtained by applying the two models to a dataset of rsfMRI images from 133 healthy subjects. FC analyses were performed using the two methods by focusing on the functional connections between the anterior part of the parahippocampal gyrus (PHGA), a core area for cognitive and emotive processes, and the rest of the brain. Based on the literature, we expect to collect evidences of the role of PHGA as connector hub between the temporal pole and the nodes of the Default Model Network. As expected, the majority of the significant links involved were highlighted by both methods. However, the FD model highlighted a greater number of links compared with pFC. These links are in line with the known neuroanatomy. Hence, our results invite to consider the FD approach as an effective approach of analysis, since due to its characteristics it could provide a more complete description of the brain network.

KW - Brain networks

KW - Default mode network

KW - Functional connectivity

KW - Functional magnetic resonance imaging

KW - Parahippocampal gyrus anterior division

KW - Resting state

KW - Temporal pole

UR - http://www.scopus.com/inward/record.url?scp=85070249098&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85070249098&partnerID=8YFLogxK

U2 - 10.1016/j.bspc.2019.101570

DO - 10.1016/j.bspc.2019.101570

M3 - Article

AN - SCOPUS:85070249098

VL - 54

JO - Biomedical Signal Processing and Control

JF - Biomedical Signal Processing and Control

SN - 1746-8094

M1 - 101570

ER -