Treatment-naïve first episode depression classification based on high-order brain functional network

Yanting Zheng, Xiaobo Chen, Danian Li, Yujie Liu, Xin Tan, Yi Liang, Han Zhang, Shijun Qiu, Dinggang Shen

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Background: Recent functional connectivity (FC) studies have proved the potential value of resting-state functional magnetic resonance imaging (rs-fMRI) in the study of major depressive disorder (MDD); yet, the rs-fMRI-based individualized diagnosis of MDD is still challenging. Methods: We enrolled 82 treatment-naïve first episode depression (FED) adults and 72 matched normal control (NC). A computer-aided diagnosis framework was utilized to classify the FEDs from the NCs based on the features extracted from not only traditional “low-order” FC networks (LON) based on temporal synchronization of original rs-fMRI signals, but also “high-order” FC networks (HON) that characterize more complex functional interactions via correlation of the dynamic (time-varying) FCs. We contrasted a classifier using HON feature (CHON) and compared its performance with using LON only (CLON). Finally, an integrated classification model with both features was proposed to further enhance FED classification. Results: The CHON had significantly improved diagnostic accuracy compared to the CLON (82.47% vs. 67.53%). Joint classification further improved the performance (83.77%). The brain regions with potential diagnostic values mainly encompass the high-order cognitive function-related networks. Importantly, we found previously less-reported potential imaging biomarkers that involve the vermis and the crus II in the cerebellum. Limitations: We only used one imaging modality and did not examine data from different subtypes of depression. Conclusions: Depression classification could be significantly improved by using HON features that better capture the higher-level brain functional interactions. The findings suggest the importance of higher-level cerebro-cerebellar interactions in the pathophysiology of MDD.

Original languageEnglish
Pages (from-to)33-41
Number of pages9
JournalJournal of Affective Disorders
Volume256
DOIs
Publication statusPublished - 2019 Sep 1

Fingerprint

Major Depressive Disorder
Depression
Magnetic Resonance Imaging
Brain
Membrane Potentials
Cerebellum
Cognition
Joints
Biomarkers

Keywords

  • Depression
  • Diagnosis
  • Dynamic functional connectivity
  • Functional magnetic resonance imaging
  • Resting state
  • Treatment naïve

ASJC Scopus subject areas

  • Clinical Psychology
  • Psychiatry and Mental health

Cite this

Treatment-naïve first episode depression classification based on high-order brain functional network. / Zheng, Yanting; Chen, Xiaobo; Li, Danian; Liu, Yujie; Tan, Xin; Liang, Yi; Zhang, Han; Qiu, Shijun; Shen, Dinggang.

In: Journal of Affective Disorders, Vol. 256, 01.09.2019, p. 33-41.

Research output: Contribution to journalArticle

Zheng, Yanting ; Chen, Xiaobo ; Li, Danian ; Liu, Yujie ; Tan, Xin ; Liang, Yi ; Zhang, Han ; Qiu, Shijun ; Shen, Dinggang. / Treatment-naïve first episode depression classification based on high-order brain functional network. In: Journal of Affective Disorders. 2019 ; Vol. 256. pp. 33-41.
@article{b1c88162e9ab4a6d939b0ff705c1f5b0,
title = "Treatment-na{\"i}ve first episode depression classification based on high-order brain functional network",
abstract = "Background: Recent functional connectivity (FC) studies have proved the potential value of resting-state functional magnetic resonance imaging (rs-fMRI) in the study of major depressive disorder (MDD); yet, the rs-fMRI-based individualized diagnosis of MDD is still challenging. Methods: We enrolled 82 treatment-na{\"i}ve first episode depression (FED) adults and 72 matched normal control (NC). A computer-aided diagnosis framework was utilized to classify the FEDs from the NCs based on the features extracted from not only traditional “low-order” FC networks (LON) based on temporal synchronization of original rs-fMRI signals, but also “high-order” FC networks (HON) that characterize more complex functional interactions via correlation of the dynamic (time-varying) FCs. We contrasted a classifier using HON feature (CHON) and compared its performance with using LON only (CLON). Finally, an integrated classification model with both features was proposed to further enhance FED classification. Results: The CHON had significantly improved diagnostic accuracy compared to the CLON (82.47{\%} vs. 67.53{\%}). Joint classification further improved the performance (83.77{\%}). The brain regions with potential diagnostic values mainly encompass the high-order cognitive function-related networks. Importantly, we found previously less-reported potential imaging biomarkers that involve the vermis and the crus II in the cerebellum. Limitations: We only used one imaging modality and did not examine data from different subtypes of depression. Conclusions: Depression classification could be significantly improved by using HON features that better capture the higher-level brain functional interactions. The findings suggest the importance of higher-level cerebro-cerebellar interactions in the pathophysiology of MDD.",
keywords = "Depression, Diagnosis, Dynamic functional connectivity, Functional magnetic resonance imaging, Resting state, Treatment na{\"i}ve",
author = "Yanting Zheng and Xiaobo Chen and Danian Li and Yujie Liu and Xin Tan and Yi Liang and Han Zhang and Shijun Qiu and Dinggang Shen",
year = "2019",
month = "9",
day = "1",
doi = "10.1016/j.jad.2019.05.067",
language = "English",
volume = "256",
pages = "33--41",
journal = "Journal of Affective Disorders",
issn = "0165-0327",
publisher = "Elsevier",

}

TY - JOUR

T1 - Treatment-naïve first episode depression classification based on high-order brain functional network

AU - Zheng, Yanting

AU - Chen, Xiaobo

AU - Li, Danian

AU - Liu, Yujie

AU - Tan, Xin

AU - Liang, Yi

AU - Zhang, Han

AU - Qiu, Shijun

AU - Shen, Dinggang

PY - 2019/9/1

Y1 - 2019/9/1

N2 - Background: Recent functional connectivity (FC) studies have proved the potential value of resting-state functional magnetic resonance imaging (rs-fMRI) in the study of major depressive disorder (MDD); yet, the rs-fMRI-based individualized diagnosis of MDD is still challenging. Methods: We enrolled 82 treatment-naïve first episode depression (FED) adults and 72 matched normal control (NC). A computer-aided diagnosis framework was utilized to classify the FEDs from the NCs based on the features extracted from not only traditional “low-order” FC networks (LON) based on temporal synchronization of original rs-fMRI signals, but also “high-order” FC networks (HON) that characterize more complex functional interactions via correlation of the dynamic (time-varying) FCs. We contrasted a classifier using HON feature (CHON) and compared its performance with using LON only (CLON). Finally, an integrated classification model with both features was proposed to further enhance FED classification. Results: The CHON had significantly improved diagnostic accuracy compared to the CLON (82.47% vs. 67.53%). Joint classification further improved the performance (83.77%). The brain regions with potential diagnostic values mainly encompass the high-order cognitive function-related networks. Importantly, we found previously less-reported potential imaging biomarkers that involve the vermis and the crus II in the cerebellum. Limitations: We only used one imaging modality and did not examine data from different subtypes of depression. Conclusions: Depression classification could be significantly improved by using HON features that better capture the higher-level brain functional interactions. The findings suggest the importance of higher-level cerebro-cerebellar interactions in the pathophysiology of MDD.

AB - Background: Recent functional connectivity (FC) studies have proved the potential value of resting-state functional magnetic resonance imaging (rs-fMRI) in the study of major depressive disorder (MDD); yet, the rs-fMRI-based individualized diagnosis of MDD is still challenging. Methods: We enrolled 82 treatment-naïve first episode depression (FED) adults and 72 matched normal control (NC). A computer-aided diagnosis framework was utilized to classify the FEDs from the NCs based on the features extracted from not only traditional “low-order” FC networks (LON) based on temporal synchronization of original rs-fMRI signals, but also “high-order” FC networks (HON) that characterize more complex functional interactions via correlation of the dynamic (time-varying) FCs. We contrasted a classifier using HON feature (CHON) and compared its performance with using LON only (CLON). Finally, an integrated classification model with both features was proposed to further enhance FED classification. Results: The CHON had significantly improved diagnostic accuracy compared to the CLON (82.47% vs. 67.53%). Joint classification further improved the performance (83.77%). The brain regions with potential diagnostic values mainly encompass the high-order cognitive function-related networks. Importantly, we found previously less-reported potential imaging biomarkers that involve the vermis and the crus II in the cerebellum. Limitations: We only used one imaging modality and did not examine data from different subtypes of depression. Conclusions: Depression classification could be significantly improved by using HON features that better capture the higher-level brain functional interactions. The findings suggest the importance of higher-level cerebro-cerebellar interactions in the pathophysiology of MDD.

KW - Depression

KW - Diagnosis

KW - Dynamic functional connectivity

KW - Functional magnetic resonance imaging

KW - Resting state

KW - Treatment naïve

UR - http://www.scopus.com/inward/record.url?scp=85067006133&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85067006133&partnerID=8YFLogxK

U2 - 10.1016/j.jad.2019.05.067

DO - 10.1016/j.jad.2019.05.067

M3 - Article

C2 - 31158714

AN - SCOPUS:85067006133

VL - 256

SP - 33

EP - 41

JO - Journal of Affective Disorders

JF - Journal of Affective Disorders

SN - 0165-0327

ER -