Multimodal hyper-connectivity of functional networks using functionally-weighted LASSO for MCI classification

Yang Li, Jingyu Liu, Xinqiang Gao, Biao Jie, Minjeong Kim, Pew Thian Yap, Chong Yaw Wee, Dinggang Shen

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Recent works have shown that hyper-networks derived from blood-oxygen-level-dependent (BOLD) fMRI, where an edge (called hyper-edge) can be connected to more than two nodes, are effective biomarkers for MCI classification. Although BOLD fMRI is a high temporal resolution fMRI approach to assess alterations in brain networks, it cannot pinpoint to a single correlation of neuronal activity since BOLD signals are composite. In contrast, arterial spin labeling (ASL) is a lower temporal resolution fMRI technique for measuring cerebral blood flow (CBF) that can provide quantitative, direct brain network physiology measurements. This paper proposes a novel sparse regression algorithm for inference of the integrated hyper-connectivity networks from BOLD fMRI and ASL fMRI. Specifically, a least absolution shrinkage and selection operator (LASSO) algorithm, which is constrained by the functional connectivity derived from ASL fMRI, is employed to estimate hyper-connectivity for characterizing BOLD-fMRI-based functional interaction among multiple regions. An ASL-derived functional connectivity is constructed by using an Ultra-GroupLASSO-UOLS algorithm, where the combination of ultra-least squares (ULS) criterion with a group LASSO (GroupLASSO) algorithm is applied to detect the topology of ASL-based functional connectivity networks, and then an ultra-orthogonal least squares (UOLS) algorithm is used to estimate the connectivity strength. By combining the complementary characterization conveyed by rs-fMRI and ASL fMRI, our multimodal hyper-networks demonstrated much better discriminative characteristics than either the conventional pairwise connectivity networks or the unimodal hyper-connectivity networks. Experimental results on publicly available ADNI dataset demonstrate that the proposed method outperforms the existing single modality based sparse functional connectivity inference methods.

Original languageEnglish
Pages (from-to)80-96
Number of pages17
JournalMedical Image Analysis
Volume52
DOIs
Publication statusPublished - 2019 Feb 1

Fingerprint

Magnetic Resonance Imaging
Labeling
Blood
Oxygen
Least-Squares Analysis
Mathematical operators
Brain
Cerebrovascular Circulation
Physiology
Biomarkers
Topology
Composite materials

Keywords

  • Arterial spin labeling (ASL)
  • Hyper-connectivity network
  • Mild cognitive impairment (MCI)
  • Multimodality
  • Ultra-least squares (ULS)
  • Weighted LASSO

ASJC Scopus subject areas

  • Radiological and Ultrasound Technology
  • Radiology Nuclear Medicine and imaging
  • Computer Vision and Pattern Recognition
  • Health Informatics
  • Computer Graphics and Computer-Aided Design

Cite this

Multimodal hyper-connectivity of functional networks using functionally-weighted LASSO for MCI classification. / Li, Yang; Liu, Jingyu; Gao, Xinqiang; Jie, Biao; Kim, Minjeong; Yap, Pew Thian; Wee, Chong Yaw; Shen, Dinggang.

In: Medical Image Analysis, Vol. 52, 01.02.2019, p. 80-96.

Research output: Contribution to journalArticle

Li, Yang ; Liu, Jingyu ; Gao, Xinqiang ; Jie, Biao ; Kim, Minjeong ; Yap, Pew Thian ; Wee, Chong Yaw ; Shen, Dinggang. / Multimodal hyper-connectivity of functional networks using functionally-weighted LASSO for MCI classification. In: Medical Image Analysis. 2019 ; Vol. 52. pp. 80-96.
@article{4cc7fe59237c4f8db3406687954dc37e,
title = "Multimodal hyper-connectivity of functional networks using functionally-weighted LASSO for MCI classification",
abstract = "Recent works have shown that hyper-networks derived from blood-oxygen-level-dependent (BOLD) fMRI, where an edge (called hyper-edge) can be connected to more than two nodes, are effective biomarkers for MCI classification. Although BOLD fMRI is a high temporal resolution fMRI approach to assess alterations in brain networks, it cannot pinpoint to a single correlation of neuronal activity since BOLD signals are composite. In contrast, arterial spin labeling (ASL) is a lower temporal resolution fMRI technique for measuring cerebral blood flow (CBF) that can provide quantitative, direct brain network physiology measurements. This paper proposes a novel sparse regression algorithm for inference of the integrated hyper-connectivity networks from BOLD fMRI and ASL fMRI. Specifically, a least absolution shrinkage and selection operator (LASSO) algorithm, which is constrained by the functional connectivity derived from ASL fMRI, is employed to estimate hyper-connectivity for characterizing BOLD-fMRI-based functional interaction among multiple regions. An ASL-derived functional connectivity is constructed by using an Ultra-GroupLASSO-UOLS algorithm, where the combination of ultra-least squares (ULS) criterion with a group LASSO (GroupLASSO) algorithm is applied to detect the topology of ASL-based functional connectivity networks, and then an ultra-orthogonal least squares (UOLS) algorithm is used to estimate the connectivity strength. By combining the complementary characterization conveyed by rs-fMRI and ASL fMRI, our multimodal hyper-networks demonstrated much better discriminative characteristics than either the conventional pairwise connectivity networks or the unimodal hyper-connectivity networks. Experimental results on publicly available ADNI dataset demonstrate that the proposed method outperforms the existing single modality based sparse functional connectivity inference methods.",
keywords = "Arterial spin labeling (ASL), Hyper-connectivity network, Mild cognitive impairment (MCI), Multimodality, Ultra-least squares (ULS), Weighted LASSO",
author = "Yang Li and Jingyu Liu and Xinqiang Gao and Biao Jie and Minjeong Kim and Yap, {Pew Thian} and Wee, {Chong Yaw} and Dinggang Shen",
year = "2019",
month = "2",
day = "1",
doi = "10.1016/j.media.2018.11.006",
language = "English",
volume = "52",
pages = "80--96",
journal = "Medical Image Analysis",
issn = "1361-8415",
publisher = "Elsevier",

}

TY - JOUR

T1 - Multimodal hyper-connectivity of functional networks using functionally-weighted LASSO for MCI classification

AU - Li, Yang

AU - Liu, Jingyu

AU - Gao, Xinqiang

AU - Jie, Biao

AU - Kim, Minjeong

AU - Yap, Pew Thian

AU - Wee, Chong Yaw

AU - Shen, Dinggang

PY - 2019/2/1

Y1 - 2019/2/1

N2 - Recent works have shown that hyper-networks derived from blood-oxygen-level-dependent (BOLD) fMRI, where an edge (called hyper-edge) can be connected to more than two nodes, are effective biomarkers for MCI classification. Although BOLD fMRI is a high temporal resolution fMRI approach to assess alterations in brain networks, it cannot pinpoint to a single correlation of neuronal activity since BOLD signals are composite. In contrast, arterial spin labeling (ASL) is a lower temporal resolution fMRI technique for measuring cerebral blood flow (CBF) that can provide quantitative, direct brain network physiology measurements. This paper proposes a novel sparse regression algorithm for inference of the integrated hyper-connectivity networks from BOLD fMRI and ASL fMRI. Specifically, a least absolution shrinkage and selection operator (LASSO) algorithm, which is constrained by the functional connectivity derived from ASL fMRI, is employed to estimate hyper-connectivity for characterizing BOLD-fMRI-based functional interaction among multiple regions. An ASL-derived functional connectivity is constructed by using an Ultra-GroupLASSO-UOLS algorithm, where the combination of ultra-least squares (ULS) criterion with a group LASSO (GroupLASSO) algorithm is applied to detect the topology of ASL-based functional connectivity networks, and then an ultra-orthogonal least squares (UOLS) algorithm is used to estimate the connectivity strength. By combining the complementary characterization conveyed by rs-fMRI and ASL fMRI, our multimodal hyper-networks demonstrated much better discriminative characteristics than either the conventional pairwise connectivity networks or the unimodal hyper-connectivity networks. Experimental results on publicly available ADNI dataset demonstrate that the proposed method outperforms the existing single modality based sparse functional connectivity inference methods.

AB - Recent works have shown that hyper-networks derived from blood-oxygen-level-dependent (BOLD) fMRI, where an edge (called hyper-edge) can be connected to more than two nodes, are effective biomarkers for MCI classification. Although BOLD fMRI is a high temporal resolution fMRI approach to assess alterations in brain networks, it cannot pinpoint to a single correlation of neuronal activity since BOLD signals are composite. In contrast, arterial spin labeling (ASL) is a lower temporal resolution fMRI technique for measuring cerebral blood flow (CBF) that can provide quantitative, direct brain network physiology measurements. This paper proposes a novel sparse regression algorithm for inference of the integrated hyper-connectivity networks from BOLD fMRI and ASL fMRI. Specifically, a least absolution shrinkage and selection operator (LASSO) algorithm, which is constrained by the functional connectivity derived from ASL fMRI, is employed to estimate hyper-connectivity for characterizing BOLD-fMRI-based functional interaction among multiple regions. An ASL-derived functional connectivity is constructed by using an Ultra-GroupLASSO-UOLS algorithm, where the combination of ultra-least squares (ULS) criterion with a group LASSO (GroupLASSO) algorithm is applied to detect the topology of ASL-based functional connectivity networks, and then an ultra-orthogonal least squares (UOLS) algorithm is used to estimate the connectivity strength. By combining the complementary characterization conveyed by rs-fMRI and ASL fMRI, our multimodal hyper-networks demonstrated much better discriminative characteristics than either the conventional pairwise connectivity networks or the unimodal hyper-connectivity networks. Experimental results on publicly available ADNI dataset demonstrate that the proposed method outperforms the existing single modality based sparse functional connectivity inference methods.

KW - Arterial spin labeling (ASL)

KW - Hyper-connectivity network

KW - Mild cognitive impairment (MCI)

KW - Multimodality

KW - Ultra-least squares (ULS)

KW - Weighted LASSO

UR - http://www.scopus.com/inward/record.url?scp=85056984235&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85056984235&partnerID=8YFLogxK

U2 - 10.1016/j.media.2018.11.006

DO - 10.1016/j.media.2018.11.006

M3 - Article

C2 - 30472348

AN - SCOPUS:85056984235

VL - 52

SP - 80

EP - 96

JO - Medical Image Analysis

JF - Medical Image Analysis

SN - 1361-8415

ER -