Identification of psychiatric disorder subtypes from functional connectivity patterns in resting-state electroencephalography

Yu, Zhang; Wu, Wei; Toll, Russell T.; Naparstek, Sharon; Maron‐Katz, Adi; Watts, Mallissa; Gordon, Joseph R.; Jeong, Jisoo; Astolfi, Laura; Shpigel, Emmanuel; Longwell, Parker; Sarhadi, Kamron; El-Said, Dawlat; Li, Yuanqing; Cooper, Crystal; Chin-Fatt, Cherise; Arns, Martijn; Goodkind, Madeleine S.; Trivedi, Madhukar H.; Marmar, Charles R.; Etkin, Amit

doi:10.1038/s41551-020-00614-8

Cited by 149 publications

(104 citation statements)

References 86 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Moreover, in our study, other regions located in the DMN, including the mPFC and LP cortices, also showed elevated activity in the restingstate, which is in line with previous findings [68]. Similarly, abnormal resting-state connectivity in the DMN has been reported in various neuropsychiatric disorders, including major depression [69], posttraumatic stress disorder (PTSD) [64], [70], and Alzheimer's [24], [71].…”

Section: B Performance Analysis Of Classification Of 3-class Sad and Hc Using Deep Learning Modelssupporting

confidence: 92%

Severity Assessment of Social Anxiety Disorder Using Deep Learning Models on Brain Effective Connectivity

et al. 2021

View full text Add to dashboard Cite

Neuroimaging investigations have proven that social anxiety disorder (SAD) is associated with aberrations in the connectivity of human brain functions.The assessment of the effective connectivity (EC) of the brain and its impact on the detection and medication of neurodegenerative pathophysiology is hence a crucial concern that needs to be addressed. Nevertheless, there are no clinically certain diagnostic biomarkers that can be linked to SAD. Therefore, investigating neural connectivity biomarkers of SAD based on deep learning models (DL) has a promising approach with its recent underlined potential results. In this study, an electroencephalography (EEG)-based detection model for SAD is constructed through directed causal influences combined with a deep convolutional neural network (CNN) and the long shortterm memory (LSTM). The EEG data were classified by applying three different DL models, namely, CNN, LSTM, and CNN+LSTM to discriminate the severity of SAD (severe, moderate, mild) and healthy controls (HC) at different frequency bands (delta, theta, alpha, low beta, and high beta) in the default mode network (DMN) under resting-state condition. The DL model uses the EC features as input, which are derived from the cortical correlation within different EEG rhythms for certain cortical areas that are more susceptible to SAD. Experimental results revealed that the proposed model (CNN+LSTM) outperforms the other models in SAD recognition. For our dataset, the highest recognition accuracies of 92.86%, 92.86%, 96.43%, and 89.29%, specificities of 95.24%, 95.24%, 100%, and 90.91%, and sensitivities of 85.71%, 85.71%, 87.50%, and 83.33% were achieved by using CNN+LSTM model for severe, moderate, mild, and HC, respectively. The fundamental contribution of this analysis is the characterization of neural brain features using different DL models to categorize the severity of SAD, which can represent a potential biomarker for SAD.INDEX TERMS Effective connectivity network, convolutional neural networks (CNNs), social anxiety disorder (SAD), default mode network (DMN), deep learning models, partial directed coherence (PDC), Electroencephalogram (EEG), human brain mapping.

show abstract

Section: B Performance Analysis Of Classification Of 3-class Sad and Hc Using Deep Learning Modelssupporting

confidence: 92%

Severity Assessment of Social Anxiety Disorder Using Deep Learning Models on Brain Effective Connectivity

et al. 2021

View full text Add to dashboard Cite

show abstract

“…The estimation of PEC between orthogonalized signals at the source-level was calculated as described in a previous study (Zhang et al, 2020). Source localization was performed via .6 ± 2.5 9.8 ± 4 9.9 ± 6.1 custom scripts and the publicly available toolbox Brainstorm (Tadel et al, 2011).…”

Section: Source-level Pec Calculationmentioning

confidence: 99%

Repetitive Transcranial Magnetic Stimulation of the Dorsolateral Prefrontal Cortex Modulates Electroencephalographic Functional Connectivity in Alzheimer’s Disease

Guo

et al. 2021

Front. Aging Neurosci.

View full text Add to dashboard Cite

Background: Increasing evidence demonstrates that repetitive transcranial magnetic stimulation (rTMS) treatment of the dorsolateral prefrontal cortex is beneficial for improving cognitive function in patients with Alzheimer’s disease (AD); however, the underlying mechanism of its therapeutic effect remains unclear.Objectives/Hypothesis: The aim of this study was to investigate the impact of rTMS to the dorsolateral prefrontal cortex on functional connectivity along with treatment response in AD patients with different severity of cognitive impairment.Methods: We conducted a 2-week treatment course of 10-Hz rTMS over the left dorsolateral prefrontal cortex in 23 patients with AD who were split into the mild or moderate cognitive impairment subgroup. Resting state electroencephalography and general cognition was assessed before and after rTMS. Power envelope connectivity was used to calculate functional connectivity at the source level. The functional connectivity of AD patients and 11 cognitively normal individuals was compared.Results: Power envelope connectivity was higher in the delta and theta bands but lower in the beta band in the moderate cognitive impairment group, compared to the cognitively normal controls, at baseline (p < 0.05). The mild cognitive impairment group had no significant abnormities. Montreal Cognitive Assessment scores improved after rTMS in the moderate and mild cognitive impairment groups. Power envelope connectivity in the beta band post-rTMS was increased in the moderate group (p < 0.05) but not in the mild group. No significant changes in the delta and theta band were found after rTMS in both the moderate and mild group.Conclusion: High-frequency rTMS to the dorsolateral prefrontal cortex modulates electroencephalographic functional connectivity while improving cognitive function in patients with AD. Increased beta connectivity may have an important mechanistic role in rTMS therapeutic effects.

show abstract

“…43 However, two recent studies using machine-learning algorithms applied to resting-state EEG features identi ed predictive signatures for sertraline, a selective serotonin-reuptake inhibitor, that related differentially to rTMS response. 44,45 This nding is of clinical relevance as it suggests that EEG signatures may be useful as a clinical tool to stratify patients to one of two evidencebased antidepressant treatments (rTMS vs. antidepressant medication), aiming to increase initial treatment response, without the requirement to consider off-label prescriptions or simply 'withhold' treatment due to a biomarker predicting low likelihood of response. 46 Here, our aim was to predict treatment outcome in MDD based on an EEG biomarker using a polygenicinformed EEG data-driven, data-reduction approach: selection of functional brain networks for subsequent response prediction was guided by PRS-MDD, thus combining genetics with neurophysiology approaches.…”

Section: Electroencephalography (Eeg) Is a Non-invasive Neuroimaging mentioning

confidence: 99%

Can polygenic-informed EEG biomarkers predict differential antidepressant treatment response? An EEG stratification marker for rTMS and sertraline

Arns

Meijs

Lin

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

The treatment of major depressive disorder (MDD) is hampered by low chances of treatment response in each treatment step, which is partly due to a lack of firmly established outcome-predictive biomarkers. Here, we hypothesize that polygenic-informed EEG biomarkers may help predict differential antidepressant treatment response. Using a polygenic-informed electroencephalography (EEG) data-driven, data-reduction approach, we identify a functional brain network that is sex-specifically associated with polygenic risk for MDD in psychiatric patients (N=1,123). Subsequently, we demonstrate the utility of this network in predicting response to transcranial magnetic stimulation (TMS) and antidepressant medication in two independent datasets (N=196 and N=1,008). A simulation aimed at stratifying patients to TMS, sertraline or escitalopram/venlafaxine based on only this EEG component yields up to >30% improved remission rates. Overall, our findings highlight the power and utility of a combined polygenic and neurophysiological approach in the search for clinically-relevant biomarkers in psychiatric disorders.

show abstract

Identification of psychiatric disorder subtypes from functional connectivity patterns in resting-state electroencephalography

Cited by 149 publications

References 86 publications

Severity Assessment of Social Anxiety Disorder Using Deep Learning Models on Brain Effective Connectivity

Severity Assessment of Social Anxiety Disorder Using Deep Learning Models on Brain Effective Connectivity

Repetitive Transcranial Magnetic Stimulation of the Dorsolateral Prefrontal Cortex Modulates Electroencephalographic Functional Connectivity in Alzheimer’s Disease

Can polygenic-informed EEG biomarkers predict differential antidepressant treatment response? An EEG stratification marker for rTMS and sertraline

Contact Info

Product

Resources

About