Comparing Cyclicity Analysis With Pre-established Functional Connectivity Methods to Identify Individuals and Subject Groups Using Resting State fMRI

Shahsavarani, Somayeh; I, Abraham; Zimmerman, Benjamin; Baryshnikov, Yuliy; Husain, Fatima T.

doi:10.3389/fncom.2019.00094

Cited by 9 publications

(15 citation statements)

References 70 publications

(122 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Global signal regression (GSR) is widely used to remove the effects of global BOLD signal variations in the analysis of fMRI studies; however, there are considerable controversies over its implementation [24][25][26], and few studies have explored the effect of GSR in ML of diagnosing neurological diseases [19,27]. Different studies have reported inconsistent results on the effect of GSR on SZ [28][29][30].…”

Section: Introductionmentioning

confidence: 99%

Machine Learning of Schizophrenia Detection with Structural and Functional Neuroimaging

Shi

Zhang

et al. 2021

Disease Markers

View full text Add to dashboard Cite

Schizophrenia (SZ) is a severe psychiatric illness, and it affects around 1% of the general population; however, its reliable diagnosis is challenging. Functional MRI (fMRI) and structural MRI (sMRI) are useful techniques for investigating the functional and structural abnormalities of the human brain, and a growing number of studies have reported that multimodal brain data can improve diagnostic accuracy. Machine learning (ML) is widely used in the diagnosis of neuroscience and neuropsychiatry diseases, and it can obtain high accuracy. However, the conventional ML which concatenated the features into a longer feature vector could not be sufficiently effective to combine different features from different modalities. There are considerable controversies over the use of global signal regression (GSR), and few studies have explored the role of GSR in ML in diagnosing neurological diseases. The current study utilized fMRI and sMRI data to implement a new method named multimodal imaging and multilevel characterization with multiclassifier (M3) to classify SZs and healthy controls (HCs) and investigate the influence of GSR in SZ classification. We found that when we used Brainnetome 246 atlas and without performed GSR, our method obtained a classification accuracy of 83.49%, with a sensitivity of 68.69%, a specificity of 93.75%, and an AUC of 0.8491, respectively. We also got great classification performances with different processing methods (with/without GSR and different brain parcellation schemes). We found that the accuracy and specificity of the models without GSR were higher than that of the models with GSR. Our findings indicate that the M3 method is an effective tool to distinguish SZs from HCs, and it can identify discriminative regions to detect SZ to explore the neural mechanisms underlying SZ. The global signal may contain important neuronal information; it can improve the accuracy and specificity of SZ detection.

show abstract

Section: Introductionmentioning

confidence: 99%

Machine Learning of Schizophrenia Detection with Structural and Functional Neuroimaging

Shi

Zhang

et al. 2021

Disease Markers

View full text Add to dashboard Cite

show abstract

“…Representative time course data from the HCP dataset after processing using Connectome Workbench (left) and corresponding lead matrix (right). The lead matrix is generated after an appropriate normalization of the BOLD signal (see [23, 25]).…”

Section: Analysis Of the Low Frequency Waves In Cortexmentioning

confidence: 99%

“…Cyclicity Analysis is a novel technique that derives pairwise temporal relations between time series using iterated path integrals (for applications in fMRI studies, see [26]). While the findings resulting from both lagged correlations and Cyclicity Analyses overlap and provide evidence in favor of the propagation of slow brain activity, they have different underlying mathematical apparatuses and assumptions, and levels of granularity.…”

Section: Cyclicity Analysismentioning

confidence: 99%

“…In the present study, using data from the Human Connectome Project [24], we ex-panded cyclicity analysis and introduced a new technique that reveals more complex dynamics of spontaneous BOLD signals. Cyclicity analysis (CA) is a novel technique that derives pairwise temporal relations between time series using iterated path integrals (for applications in fMRI studies, see [25]). While the findings resulting from both lagged correlations and cyclicity analyses overlap and provide evidence in favor of the propagation of slow brain activity, they have different underlying mathematical apparatuses and assumptions, and levels of granularity.…”

Section: Introductionmentioning

confidence: 99%

“…The lagged correlation method infers lag threads by deriving singular vectors of the time-delay matrix, whereas the CA method recovers inherent ordering among BOLD time series through eigenvectors of a lead matrix (a representation of the strength of temporal ordering between pairs of regions, see Section 2). Moreover, lagged correlation relies on interpolation and windowing to capture the dynamics of FC, which is vulnerable to time delay estimation methods, autocorrelation [25], sampling variability [26], and parameters such as window length and window shift [27]. In contrast, cyclicity analysis offers a more robust approach with a higher level of granularity to study lag structure where there is no assumption regarding stationarity, latency estimation, state duration, and state transition.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Slow Cortical Waves through Cyclicity Analysis

Shahsavarani²,

Zimmerman

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Fine-grained information about dynamic structure of cortical networks is crucial in unpacking brain function. Here,we introduced a novel analytical method to characterize the dynamic interaction between distant brain regions,based on cyclicity analysis, and applied it to data from the Human Connectome Project. Resting-state fMRI time series are aperiodic and, hence, lack a base frequency. Cyclicity analysis, which is time-reparametrization invariant, is effective in recovering dynamic temporal ordering of such time series along a circular trajectory without assuming any time scale. Our analysis detected the propagation of slow cortical waves across thebrain with consistent shifts in lead-lag relationships between specific brain regions. We also observed short bursts of strong temporal ordering that dominated overall lead-lag relationships between pairs of regions in the brain, which were modulated by tasks. Our results suggest the possible role played by slow waves of ordered information between brain regions that underlie emergent cognitive function.

show abstract

Salience, emotion, and attention: The neural networks underlying tinnitus distress revealed using music and rest

et al. 2021

View full text Add to dashboard Cite

Comparing Cyclicity Analysis With Pre-established Functional Connectivity Methods to Identify Individuals and Subject Groups Using Resting State fMRI

Cited by 9 publications

References 70 publications

Machine Learning of Schizophrenia Detection with Structural and Functional Neuroimaging

Machine Learning of Schizophrenia Detection with Structural and Functional Neuroimaging

Slow Cortical Waves through Cyclicity Analysis

Salience, emotion, and attention: The neural networks underlying tinnitus distress revealed using music and rest

Contact Info

Product

Resources

About