Estimating Dynamic Functional Brain Connectivity With a Sparse Hidden Markov Model

Zhang, Gemeng; Cai, Biao; Zhang, Aiying; Stephen, Julia M.; Wilson, Tony W.; Calhoun, Vince D.; Wang, Yu Ping

doi:10.1109/tmi.2019.2929959

Cited by 46 publications

(20 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…HMM can be viewed as a probabilistic description of the data in which the hidden information (i.e., brain state) is captured from observable measures (i.e., fMRI time courses). An increasing number of studies have applied HMM to neuroimaging data, demonstrating it as a promising approach for understanding the temporal dynamics of brain functions [52][53][54][55] .…”

Section: Dynamic Brain Statesmentioning

confidence: 99%

Mental health explains individual deviations from normative range in cognition-associated brain states

Zhang

Vidaurre

Bijsterbosch

2021

Preprint

View full text Add to dashboard Cite

Links between cognitive deficits and psychiatric disorders have been studied predominantly at the group level, leaving unique individual characteristics largely unknown. Here, we applied normative modeling to UK Biobank data (N=18,634) and estimated the interplay of large-scale brain networks over time (i.e., dynamic brain state) at the individual level. Abnormality in such brain states was linked to individual variation in mental health. Specifically, brain state measures including fractional occupancy were estimated as a function of general cognitive ability and abnormality scores per participant were quantified to represent the degree of deviations relative to the estimated population norm. We found significant associations between the abnormality scores of several brain states and individuals overall mental health. Our findings suggest potential impact of mental health on dynamic brain states subserving cognitive functions and shed light on the relevant brain mechanisms underlying cognitive deficits in mental illness.

show abstract

Section: Dynamic Brain Statesmentioning

confidence: 99%

Mental health explains individual deviations from normative range in cognition-associated brain states

Zhang

Vidaurre

Bijsterbosch

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…In addition, Amico et al showed that the individual fingerprint of a human functional connectome could be improved from a reconstruction procedure based on a group‐wise decomposition (Amico & Goñi, 2018). Recently, the limitation of static connectivity has been widely realized, and the concept of dynamic connectivity has emerged to emphasize the time‐varying characteristics of the FC (Allen et al, 2014; Cai et al, 2017, 2018; Calhoun & Adali, 2016; Calhoun, Miller, Pearlson, & Adalı, 2014; Hutchison et al, 2013; Hutchison, Womelsdorf, Gati, Everling, & Menon, 2013; Zhang et al, 2019; Zhang et al, 2019; Zhang, Fang, Liang, Calhoun, & Wang, 2018). Incorporating the information from the time‐varying FC, Liu et al studied whether and how the dynamic properties of the chronnectome acted as a fingerprint of the brain to identify individuals (Liu, Liao, Xia, & He, 2018).…”

Section: Introductionmentioning

confidence: 99%

Functional connectome fingerprinting: Identifying individuals and predicting cognitive functions via autoencoder

Cai

Zhang

et al. 2021

Human Brain Mapping

Self Cite

View full text Add to dashboard Cite

Functional network connectivity has been widely acknowledged to characterize brain functions, which can be regarded as “brain fingerprinting” to identify an individual from a pool of subjects. Both common and unique information has been shown to exist in the connectomes across individuals. However, very little is known about whether and how this information can be used to predict the individual variability of the brain. In this paper, we propose to enhance the uniqueness of individual connectome based on an autoencoder network. Specifically, we hypothesize that the common neural activities shared across individuals may reduce the individual identification. By removing contributions from shared activities, inter‐subject variability can be enhanced. Our experimental results on HCP data show that the refined connectomes obtained by utilizing autoencoder with sparse dictionary learning can distinguish an individual from the remaining participants with high accuracy (up to 99.5% for the rest–rest pair). Furthermore, high‐level cognitive behaviors (e.g., fluid intelligence, executive function, and language comprehension) can also be better predicted with the obtained refined connectomes. We also find that high‐order association cortices contribute more to both individual discrimination and behavior prediction. In summary, our proposed framework provides a promising way to leverage functional connectivity networks for cognition and behavior study, in addition to a better understanding of brain functions.

show abstract

“…To measure time-varying fMRI FC, we used a state-based model where each state is associated with a specific pattern of FC (Vidaurre et al, 2017), such that instantaneous changes in FC manifest as a change of state. This approach is based on a version of the hidden Markov model (HMM) that, in comparison to previous versions of the HMM used on fMRI (Vidaurre et al, 2017; Stevner et al, 2019; Baldassano et al, 2017; Shappell et al, 2019; Zhang et al, 2019), emphasises changes in FC over changes in amplitude. To model each subject, the HMM uses a temporally-organised mixture of (quasi-) stable FC patterns in the form of region-by-region covariance matrices.…”

Section: Introductionmentioning

confidence: 99%

Behavioural relevance of spontaneous, transient brain network interactions in fMRI

Vidaurre¹,

Llera²,

Smith³

et al. 2019

Preprint

View full text Add to dashboard Cite

How spontaneously fluctuating functional magnetic resonance imaging (fMRI) signals in different brain regions relate to behaviour has been an open question for decades. Correlations in these signals, known as functional connectivity, can be averaged over several minutes of data to provide a stable representation of the functional network architecture for an individual. However, associations between these stable features and behavioural traits have been shown to be dominated by individual differences in anatomy. Here, we show that timevarying fMRI functional connectivity, detected at time-scales of several seconds, has associations with behavioural traits that are not dominated by anatomy. We found that this was the case for several trait groups, despite the fact that average functional connectivity accounts for the largest proportion of variability in the fMRI signal between individuals. The finding that time-varying fMRI functional connectivity has a unique relationship to population behavioural variability implies that it might reflect transient neuronal communication fluctuating around a stable neural architecture.normally subtle effects, and other researchers have reported little or no differences in FC between task and rest (Hampson et al., 2006;Gratton et al., 2018).Here, we take a different route, by relating time-varying FC to population variability in behavioural traits. For this purpose, we combined subject-specific estimates of time-varying fMRI FC, average FC and structural connectivity -for the first time considering these three elements in one single analysis. Using different groups of behavioural traits, we explored the relationship between time-varying FC and population behaviour, after accounting for the explanatory power of the stable subject-specific average FC and structural connectivity features. We reasoned that if time-varying fMRI FC represents biologically meaningful communication between neuronal populations, then it should be capable of accounting for aspects of the subjects' behavioural phenotypes not explained by the average FC or the structural connectivity information. We also examined the relationship between time-varying FC and a separate group of anatomical traits, further confirming that time-varying FC was fundamentally unrelated to the anatomy.

show abstract

Estimating Dynamic Functional Brain Connectivity With a Sparse Hidden Markov Model

Cited by 46 publications

References 51 publications

Mental health explains individual deviations from normative range in cognition-associated brain states

Mental health explains individual deviations from normative range in cognition-associated brain states

Functional connectome fingerprinting: Identifying individuals and predicting cognitive functions via autoencoder

Behavioural relevance of spontaneous, transient brain network interactions in fMRI

Contact Info

Product

Resources

About