Discriminative sparse connectivity patterns for classification of fMRI Data.

Harini Eavani, Theodore D. Satterthwaite, Raquel E. Gur, Ruben C. Gur, Christos Davatzikos

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

Functional connectivity using resting-state fMRI has emerged as an important research tool for understanding normal brain function as well as changes occurring during brain development and in various brain disorders. Most prior work has examined changes in pairwise functional connectivity values using a multi-variate classification approach, such as Support Vector Machines (SVM). While it is powerful, SVMs produce a dense set of high-dimensional weight vectors as output, which are difficult to interpret, and require additional post-processing to relate to known functional networks. In this paper, we propose a joint framework that combines network identification and classification, resulting in a set of networks, or Sparse Connectivity Patterns (SCPs) which are functionally interpretable as well as highly discriminative of the two groups. Applied to a study of normal development classifying children vs. adults, the proposed method provided accuracy of 76%(AUC= 0.85), comparable to SVM (79%,AUC=0.87), but with dramatically fewer number of features (50 features vs. 34716 for the SVM). More importantly, this leads to a tremendous improvement in neuro-scientific interpretability, which is specially advantageous in such a study where the group differences are wide-spread throughout the brain. Highest-ranked discriminative SCPs reflect increases in long-range connectivity in adults between the frontal areas and posterior cingulate regions. In contrast, connectivity between the bilateral parahippocampal gyri was decreased in adults compared to children.

Original languageEnglish (US)
Title of host publicationMedical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention
Pages193-200
Number of pages8
Volume17
EditionPt 3
StatePublished - 2014
Externally publishedYes

Fingerprint

Magnetic Resonance Imaging
Area Under Curve
Brain
Parahippocampal Gyrus
Gyrus Cinguli
Brain Diseases
Child Development
Joints
Weights and Measures
Research
Support Vector Machine

ASJC Scopus subject areas

  • Medicine(all)

Cite this

Eavani, H., Satterthwaite, T. D., Gur, R. E., Gur, R. C., & Davatzikos, C. (2014). Discriminative sparse connectivity patterns for classification of fMRI Data. In Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention (Pt 3 ed., Vol. 17, pp. 193-200)

Discriminative sparse connectivity patterns for classification of fMRI Data. / Eavani, Harini; Satterthwaite, Theodore D.; Gur, Raquel E.; Gur, Ruben C.; Davatzikos, Christos.

Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention. Vol. 17 Pt 3. ed. 2014. p. 193-200.

Research output: Chapter in Book/Report/Conference proceedingChapter

Eavani, H, Satterthwaite, TD, Gur, RE, Gur, RC & Davatzikos, C 2014, Discriminative sparse connectivity patterns for classification of fMRI Data. in Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention. Pt 3 edn, vol. 17, pp. 193-200.
Eavani H, Satterthwaite TD, Gur RE, Gur RC, Davatzikos C. Discriminative sparse connectivity patterns for classification of fMRI Data. In Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention. Pt 3 ed. Vol. 17. 2014. p. 193-200
Eavani, Harini ; Satterthwaite, Theodore D. ; Gur, Raquel E. ; Gur, Ruben C. ; Davatzikos, Christos. / Discriminative sparse connectivity patterns for classification of fMRI Data. Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention. Vol. 17 Pt 3. ed. 2014. pp. 193-200
@inbook{ec8b0690ea944a63b7eace3f218711c0,
title = "Discriminative sparse connectivity patterns for classification of fMRI Data.",
abstract = "Functional connectivity using resting-state fMRI has emerged as an important research tool for understanding normal brain function as well as changes occurring during brain development and in various brain disorders. Most prior work has examined changes in pairwise functional connectivity values using a multi-variate classification approach, such as Support Vector Machines (SVM). While it is powerful, SVMs produce a dense set of high-dimensional weight vectors as output, which are difficult to interpret, and require additional post-processing to relate to known functional networks. In this paper, we propose a joint framework that combines network identification and classification, resulting in a set of networks, or Sparse Connectivity Patterns (SCPs) which are functionally interpretable as well as highly discriminative of the two groups. Applied to a study of normal development classifying children vs. adults, the proposed method provided accuracy of 76{\%}(AUC= 0.85), comparable to SVM (79{\%},AUC=0.87), but with dramatically fewer number of features (50 features vs. 34716 for the SVM). More importantly, this leads to a tremendous improvement in neuro-scientific interpretability, which is specially advantageous in such a study where the group differences are wide-spread throughout the brain. Highest-ranked discriminative SCPs reflect increases in long-range connectivity in adults between the frontal areas and posterior cingulate regions. In contrast, connectivity between the bilateral parahippocampal gyri was decreased in adults compared to children.",
author = "Harini Eavani and Satterthwaite, {Theodore D.} and Gur, {Raquel E.} and Gur, {Ruben C.} and Christos Davatzikos",
year = "2014",
language = "English (US)",
volume = "17",
pages = "193--200",
booktitle = "Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention",
edition = "Pt 3",

}

TY - CHAP

T1 - Discriminative sparse connectivity patterns for classification of fMRI Data.

AU - Eavani, Harini

AU - Satterthwaite, Theodore D.

AU - Gur, Raquel E.

AU - Gur, Ruben C.

AU - Davatzikos, Christos

PY - 2014

Y1 - 2014

N2 - Functional connectivity using resting-state fMRI has emerged as an important research tool for understanding normal brain function as well as changes occurring during brain development and in various brain disorders. Most prior work has examined changes in pairwise functional connectivity values using a multi-variate classification approach, such as Support Vector Machines (SVM). While it is powerful, SVMs produce a dense set of high-dimensional weight vectors as output, which are difficult to interpret, and require additional post-processing to relate to known functional networks. In this paper, we propose a joint framework that combines network identification and classification, resulting in a set of networks, or Sparse Connectivity Patterns (SCPs) which are functionally interpretable as well as highly discriminative of the two groups. Applied to a study of normal development classifying children vs. adults, the proposed method provided accuracy of 76%(AUC= 0.85), comparable to SVM (79%,AUC=0.87), but with dramatically fewer number of features (50 features vs. 34716 for the SVM). More importantly, this leads to a tremendous improvement in neuro-scientific interpretability, which is specially advantageous in such a study where the group differences are wide-spread throughout the brain. Highest-ranked discriminative SCPs reflect increases in long-range connectivity in adults between the frontal areas and posterior cingulate regions. In contrast, connectivity between the bilateral parahippocampal gyri was decreased in adults compared to children.

AB - Functional connectivity using resting-state fMRI has emerged as an important research tool for understanding normal brain function as well as changes occurring during brain development and in various brain disorders. Most prior work has examined changes in pairwise functional connectivity values using a multi-variate classification approach, such as Support Vector Machines (SVM). While it is powerful, SVMs produce a dense set of high-dimensional weight vectors as output, which are difficult to interpret, and require additional post-processing to relate to known functional networks. In this paper, we propose a joint framework that combines network identification and classification, resulting in a set of networks, or Sparse Connectivity Patterns (SCPs) which are functionally interpretable as well as highly discriminative of the two groups. Applied to a study of normal development classifying children vs. adults, the proposed method provided accuracy of 76%(AUC= 0.85), comparable to SVM (79%,AUC=0.87), but with dramatically fewer number of features (50 features vs. 34716 for the SVM). More importantly, this leads to a tremendous improvement in neuro-scientific interpretability, which is specially advantageous in such a study where the group differences are wide-spread throughout the brain. Highest-ranked discriminative SCPs reflect increases in long-range connectivity in adults between the frontal areas and posterior cingulate regions. In contrast, connectivity between the bilateral parahippocampal gyri was decreased in adults compared to children.

UR - http://www.scopus.com/inward/record.url?scp=84909642457&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84909642457&partnerID=8YFLogxK

M3 - Chapter

VL - 17

SP - 193

EP - 200

BT - Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention

ER -