Evaluating functional connectivity of executive control network and frontoparietal network in Alzheimer’s disease

Zhao, Qinghua; Lü, Hong; Metmer, Hichem; Li, Will X. Y.; Lu, Jianfeng

doi:10.1016/j.brainres.2017.10.025

Cited by 60 publications

(34 citation statements)

References 57 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While promising results based on group-level comparisons were found [2–4]; computer-aided individualized eMCI diagnosis is still among the most challenging tasks. Many MCI studies have used resting-state functional Magnetic Resonance Imaging (rs-fMRI)-derived brain functional networks (BFNs) for MCI studies [5, 6]; however, most of them are based on mass-univariate analysis, e.g., constructing the default mode network (DMN) and conducting a voxel-wise greedy search for potential group differences within it [5]. While such traditional BFN-based MCI studies yield some statistically significant brain regions [2–4], they suffer several inherent drawbacks: ( 1 ) mass-univariate analysis usually results in loss of statistical power; ( 2 ) the voxel-wise analysis may lose rich inter-voxel pattern information in the BFNs; and ( 3 ) the severe noise in the rs-fMRI data could overwhelm subtle diagnostic information in eMCI.…”

Section: Introductionmentioning

confidence: 99%

A Novel Deep Learning Framework on Brain Functional Networks for Early MCI Diagnosis

Kam

Zhang

Shen

2018

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Although alternations of brain functional networks (BFNs) derived from resting-state functional magnetic resonance imaging (rs-fMRI) have been considered as promising biomarkers for early Alzheimer’s disease (AD) diagnosis, it is still challenging to perform individualized diagnosis, especially at the very early stage of preclinical stage of AD, i.e., early mild cognitive impairment (eMCI). Recently, convolutional neural networks (CNNs) show powerful ability in computer vision and image analysis applications, but there is still a gap for directly applying CNNs to rs-fMRI-based disease diagnosis. In this paper, we propose a novel multiple-BFN-based 3D CNN framework that can automatically and deeply learn complex, high-level, hierarchical diagnostic features from various independent component analysis-derived BFNs. More importantly, the embedded features of different BFNs could comprehensively support each other towards a more accurate eMCI diagnosis in a unified model. The performance of the proposed method is validated by a large-sample, multisite, rigorously controlled publicly accessible dataset. The proposed framework can also be conveniently and straightforwardly applied to individualized diagnosis of various neurological and psychiatric diseases.

show abstract

Section: Introductionmentioning

confidence: 99%

A Novel Deep Learning Framework on Brain Functional Networks for Early MCI Diagnosis

Kam

Zhang

Shen

2018

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…This state is characterized by positive connections between DMN and attention network and SN. This state contained FC between brain regions where AD dysconnectivity have been repeatedly reported in previous neuroimaging studies, including the posterior cingulate/precuneus (Sorg et al, 2009; Yokoi et al, 2018), anterior cingulate cortex (Amanzio et al, 2011; Brier et al, 2012), medial temporal cortex (Pasquini et al, 2019; Sorg et al, 2009), and fontal cortex (Zhao et al, 2018). Interestingly, the correlated patterns in FPN-SN-DMN and SMN-sub-VIS states were anti-correlated which indicated that AD patients significantly differ from HC subjects in probability of occurrence of these anti-correlated FC patterns.…”

Section: Discussionmentioning

confidence: 94%

“…Disruption of several resting-state networks in AD have been reported, including the default mode network (DMN), involved in memory formation and retrieval (Andrews-Hanna, 2012) and self-related processing (L. Zhang et al, 2020), frontoparietal network (FPN), flexibly controlling and interacting with other brain networks (Marek & Dosenbach, 2018), the salience network (SN), directing attention toward or away from internal processing in concert with the DMN (Menon & Uddin, 2010), somatomotor network (SMN) and visual network (Mosimann, Felblinger, Ballinari, Hess, & Muri, 2004). Previous studies have ignored dynamic nature of FC and mostly focused on assessing ‘static’ FC which represents mean connectivity over the period of scanning (Damoiseaux, Prater, Miller, & Greicius, 2012; Greicius, Srivastava, Reiss, & Menon, 2004; Sorg, Riedl, Perneczky, Kurz, & Wohlschlager, 2009; Zhao, Lu, Metmer, Li, & Lu, 2018). Most recently, several studies have provided empirical evidences in healthy subjects (Cabral et al, 2017; Larabi et al, 2020; Maleki Balajoo, Asemani, Khadem, & Soltanian-Zadeh, 2020) and psychiatric (Figueroa et al, 2019; Sakoglu et al, 2010) and neurological (Fu et al, 2019; Gu et al, 2020; Jones et al, 2012; Kim et al, 2017; Niu et al, 2019; Schumacher et al, 2019; Sourty et al, 2016) disorders that not only the intrinsic brain FC at rest is dynamic during the period of scanning (Allen et al, 2014; Betzel, Fukushima, He, Zuo, & Sporns, 2016; Chang & Glover, 2010; Hutchison et al, 2013), but also temporal properties of FC reconfiguration over time are associated with symptoms, behavioral and cognitive performance (Cabral et al, 2017; Figueroa et al, 2019; Gu et al, 2020; Larabi et al, 2020; Tian, Li, Wang, & Yu, 2018; Viviano, Raz, Yuan, & Damoiseaux, 2017).…”

Section: Introductionmentioning

confidence: 99%

Impaired dynamic profile of brain recurring functional connectivity states in Alzheimer’s disease

Balajoo¹,

Asemani²,

Khadem³

et al. 2020

Preprint

View full text Add to dashboard Cite

Up to now, disrupted functional organization of the brain in Alzheimer’s disease (AD) has mostly been examined by assessing “static” functional connectivity (FC). However, recent studies have provided clear evidence that brain FC exhibits intrinsic spatiotemporal “dynamic” organization also being significantly affected in AD. However, inconsistency in former studies is a motivation for further investigation of impaired dynamic profile of brain FC networks in Alzheimer’s disease._In this work, we identified recurring FC states during rest (eyes-open) in 24 AD patients and 37 healthy controls (HC) by combining sliding window-based method (examining dynamic FC) and k-means clustering algorithm (identifying recurring FC states over time). To define the optimum number of recurring FC states in k-means algorithm, we employed non-supervised validity criteria such as silhouette value and Dunn index over bootstrap samples. Then, we examined differences in dynamic properties of recurring FC states including probability of occurrence, lifetime and switching profile between groups. Afterward, association between impaired dynamic profile of recurring FC states and cognitive performance was assessed in AD patients. Our findings revealed three recurring FC states and among them, FC state with the most significant differences in probability of occurrence (corrected p-value < 0.0001) included brain regions involved in default mode, frontoparietal and salience networks. This recurring FC state occurred significantly less and lasted shorter in patients with AD than in the HC. Further, the probability of occurrence in this recurring FC state significantly correlated positively with cognitive decline in patients with AD. Our findings suggest more decreased in cognitive performance in patients with AD associated with more reduced ability of AD patients to access clinically relevant impaired brain FC networks. Impaired dynamic profile of recurring FC states in AD provide greater understanding of AD’s pathophysiology.

show abstract

“…Reduced parietal activation underpinned the change in topography of class D in AD, supporting the hypothesis of dysfunction of the frontoparietal network. A possible mechanism for this is disrupted frontoparietal white matter integrity, since altered frontoparietal functional and effective connectivity have been reported in recent fMRI studies of AD (Neufang et al, 2011;Zhao et al, 2018).…”

Section: Alterations To Class D and The Frontoparietal Networkmentioning

confidence: 99%

EEG microstate complexity for aiding early diagnosis of Alzheimer’s disease

Tait

Tamagnini

Stothart

et al. 2019

Preprint

View full text Add to dashboard Cite

Introduction: Electroencephalogram (EEG) is a potentially useful clinical tool for aiding diagnosis of Alzheimer's disease (AD). We hypothesized we can increase the accuracy of EEG for aiding diagnosis of AD using microstates, which are epochs of quasi-stability at the millisecond scale.Methods: EEG was collected from two independent cohorts of AD and control participants and a cohort of mild cognitive impairment (MCI) patients with four-year clinical follow-up. Microstates were analysed, including a novel measure of complexity.Results: Microstate complexity significantly decreased in AD, and when combined with a spectral EEG measure, could classify AD with sensitivity and specificity >80%. These results were validated on an independent cohort and were also found to be generalizable to predict progression from MCI to AD. Additionally, microstates associated with the frontoparietal network were altered in AD.Discussion: EEG has the potential to be a non-invasive functional biomarker that predicts progression from MCI to AD.

show abstract

Evaluating functional connectivity of executive control network and frontoparietal network in Alzheimer’s disease

Cited by 60 publications

References 57 publications

A Novel Deep Learning Framework on Brain Functional Networks for Early MCI Diagnosis

A Novel Deep Learning Framework on Brain Functional Networks for Early MCI Diagnosis

Impaired dynamic profile of brain recurring functional connectivity states in Alzheimer’s disease

EEG microstate complexity for aiding early diagnosis of Alzheimer’s disease

Contact Info

Product

Resources

About