Temporal Frequency Patterns Of Magnetoencephalography For The Recognition Of Bipolar Disorder Under Emotional Face Stimulation

Bipolar disorder(BD)without experiencing a manic episode and in depressed state is often misdiagnosed as unipolar depression(UD),and timely diagnosis and accurate treatment are crucial for BD patients.Functional impairment and abnormal regulation of mood-related networks have been found in BD patients,but the time-frequency specificity of the relevant brain networks in response to negative emotional stimuli needs further investigation.In this paper,we explored the imaging markers that distinguish BD patients who do not experience manic episodes and are in depressive phase(t BD)from depression by using the high temporal and spatial resolution and multi-band properties of magnetoencephalography(MEG)in conjunction with the sad face emotion stimuli paradigm.By exploring the specificity of brain networks in t BD,it can assist BD patients in the early clinical diagnosis.The main contents of this paper are as follows:1.Exploring differences in brain network patterns of BD,t BD,UD and healthy controls(HC)after sad face emotion stimuli.The experimental samples consisted of 27 BD,24 t BD,30 UD and 26 HC acquired by MEG,extracting low-gamma band(30–50 Hz)signals for 600 ms after sad face emotion stimuli,and divided into three time periods.First,network-based statistics was used to explore the differences between diseases in the depressive phase(t BD and UD)and HC.Then,the two-sample t-test was used to explore the changes in the brain network of BD patients after manic episodes.It was found that the strength of the PreCG.L-VMPFC connection(left precentral gyrus and ventral medial prefrontal cortex)was able to distinguish between t BD,UD and HC in the early time period(0–200 ms)and BD patients with and without manic episode in the middle time period(200–400 ms).We also found that the ratio of the strength of the PreCG.L-VMPFC connection in the middle/early time period was significantly negatively correlated with the bipolarity index,which may reflect the degree of impairment of negative emotion processing in BD.2.The signal modulation direction of differential connectivity in BD,t BD,UD and HC was further analyzed after sad face emotion stimuli.The experimental subjects were the same as above,with the difference being the acquisition of the full frequency band(1-80 Hz)MEG signal 600 ms after stimulation.We analyzed the direction of differentially connected signal modulation using both the phase slope index(PSI)and the peak latency(PL),respectively.We used a one-way ANOVA to count differences in the direction of signal modulation between disease in depressive phase(t BD and UD)and HC.The two-sample t-test was used to explore changes in the direction of signal modulation in BD patients after manic episodes.The results revealed a bottom-up modulation of the PreCG.L-VMPFC connection in the higher frequency band(around 70 Hz).This may be a unique biomarker of phase inversion from typical BD,UD patients and HC.At 75–78 Hz in the ITG.L-DMPFC.L connection(left inferior temporal gyrus and left dorsomedial prefrontal cortex),we found opposite direction of signal modulation in t BD and BD patients,which could be an impairment caused by manic episodes.3.Temporal variability in the connectivity strength of stimulus-related connections of sademotional faces was analyzed in BD,t BD,UD,and HC.The experimental subjects were the same as above,with the difference being the acquisition of the full frequency band(1–80 Hz)MEG signal 1000 ms after stimulation and divided into six frequency bands.For each band of signals,the brain regions of interest that appear connected in each group after sad face emotion stimuli are adaptively selected by an adaptive least absolute shrinkage and selection operator(LASSO)algorithm.The recursive least squares(RLS)algorithm calculateed the connection strength of these brain regions at each time sampling point.The statistical method is the same as in the previous section.It was found that the dynamic functional connectivity between 200-600 ms after stimulation could distinguish the t BD,and the mean dynamic functional connectivity value of the t BD after 600 ms was correlated with the bipolarity index.The connection of the low-gamma band can distinguishes the t BD and verifies the correctness of the previous selection of the low-gamma band.The salience network(SN)in bipolar disorder may be influenced by manic strikes,reflecting differences from UD.However,due to the small sample size and individual differences among subjects in this study,as well as the differences in feature extraction methods,the same connection as in the previous paper was not found.More subjects can be included in subsequent studies to reduce individual differences.In summary,the low-gamma frequency band after sad face emotion stimuli is a key frequency band for distinguishing t BD patients,and the PreCG.L-VMPFC connection is associated with manic episodes in BD,We further explored the temporal segments and brain networks that are useful for differentiating t BD patients in terms of temporal variability.