Explore the brain connectivity of panic disorder patients based on the graph theoretical approach
Poster No:
695
Submission Type:
Abstract Submission
Authors:
Hye Jin Hong1, Ji Seon Ahn2,3,4, Jin Young Park2,3,4, Jee Hang Lee1,5
Institutions:
1Department of AI & Informatics, Sangmyung University, Seoul, Korea, Republic of, 2Department of Psychiatry, Yonsei University College of Medicine, Yongin Severance Hospital, Yongin, Korea, Republic of, 3Institute of Behavioral Science in Medicine, Yonsei University College of Medicine, Seoul, Korea, Republic of, 4Center for Digital Health, Yongin Severance Hospital, Yonsei University Health System, Yongin, Korea, Republic of, 5Department of Human-Centered AI, Sangmyung University, Seoul, Korea, Republic of
First Author:
Co-Author(s):
Ji Seon Ahn
Department of Psychiatry, Yonsei University College of Medicine, Yongin Severance Hospital|Institute of Behavioral Science in Medicine, Yonsei University College of Medicine|Center for Digital Health, Yongin Severance Hospital, Yonsei University Health System
Yongin, Korea, Republic of|Seoul, Korea, Republic of|Yongin, Korea, Republic of
Department of Psychiatry, Yonsei University College of Medicine, Yongin Severance Hospital|Institute of Behavioral Science in Medicine, Yonsei University College of Medicine|Center for Digital Health, Yongin Severance Hospital, Yonsei University Health System
Yongin, Korea, Republic of|Seoul, Korea, Republic of|Yongin, Korea, Republic of
Jin Young Park
Department of Psychiatry, Yonsei University College of Medicine, Yongin Severance Hospital|Institute of Behavioral Science in Medicine, Yonsei University College of Medicine|Center for Digital Health, Yongin Severance Hospital, Yonsei University Health System
Yongin, Korea, Republic of|Seoul, Korea, Republic of|Yongin, Korea, Republic of
Department of Psychiatry, Yonsei University College of Medicine, Yongin Severance Hospital|Institute of Behavioral Science in Medicine, Yonsei University College of Medicine|Center for Digital Health, Yongin Severance Hospital, Yonsei University Health System
Yongin, Korea, Republic of|Seoul, Korea, Republic of|Yongin, Korea, Republic of
Jee Hang Lee
Department of AI & Informatics, Sangmyung University|Department of Human-Centered AI, Sangmyung University
Seoul, Korea, Republic of|Seoul, Korea, Republic of
Department of AI & Informatics, Sangmyung University|Department of Human-Centered AI, Sangmyung University
Seoul, Korea, Republic of|Seoul, Korea, Republic of
Introduction:
It is widely accepted that the brain structure and functions of patients with panic disorder (PD) are different from those of healthy controls (HCs) (Imperatori et al., 2019). In this study, we would like to explore the discrepancy of brain connectivity between patients with panic disorder and healthy control based on the graph theoretical approach using EEG. To that end, we analyzed EEG data during both resting state (RS) and mental arithmetic (MA) tasks to explore the topological characteristics of the brain associated with information processing difficulties, particularly cognitive challenges observed in individuals with panic disorder. To construct functional networks within the individuals with PD and HC, we adopted a network perspective, treating the brain as a complex network. Employing a graph theoretical approach, a widely recognized means for assessing information communication in the human brain, we aimed to unveil topological alterations indicative of panic disorder-related changes in information processing dynamics.
Methods:
In this study, 34 participants were included as a PD patients group, who had been diagnosed with PD. To recruit them, we retrospectively analyzed medical records and EEG data from patients who sought treatment for anxiety at a psychiatric outpatient clinic between Mar. 1, 2020 and Sep. 30, 2023. Paired with the PD group, we recruited the additional 34 healthy controls matched with respect to age and gender distribution (IRB reference number 9-2022-0199, Feb. 20, 2023).
The experiment consisted of two sessions. In the first, participants were asked to keep their eyes closed for five minutes (we called the resting state; RS). Next, participants were asked to perform the mental arithmetic task for five minutes while keeping eyes closed (we called the mental arithmetic;MA). EEG data were recorded in both sessions.
We used the graph-theoretical approach, specifically, computed Phase Locking Value (PLV) (Lachaux et al., 1999) among all pairs of EEG channels. It quantifies the degree of phase synchronization between two narrow-band signals (Aydore et al., 2013). We then constructed the brain network using the PLV matrices whose values were above 25 percentile. For each of the reconstructed graphs, two indices were computed: (i) Global Efficiency (GE) (Latora & Marchiori, 2001); (ii) Local Efficiency (LE) (Latora & Marchiori, 2001).
The experiment consisted of two sessions. In the first, participants were asked to keep their eyes closed for five minutes (we called the resting state; RS). Next, participants were asked to perform the mental arithmetic task for five minutes while keeping eyes closed (we called the mental arithmetic;MA). EEG data were recorded in both sessions.
We used the graph-theoretical approach, specifically, computed Phase Locking Value (PLV) (Lachaux et al., 1999) among all pairs of EEG channels. It quantifies the degree of phase synchronization between two narrow-band signals (Aydore et al., 2013). We then constructed the brain network using the PLV matrices whose values were above 25 percentile. For each of the reconstructed graphs, two indices were computed: (i) Global Efficiency (GE) (Latora & Marchiori, 2001); (ii) Local Efficiency (LE) (Latora & Marchiori, 2001).
Results:
Comparing the two graph network measures for all frequency bands between PD and HC in a RS, GE of the theta, beta, and gamma band was significantly higher in the HC group than that in the PD group while no significant results were found in the rest (Figure 1A-left). On the other hand, for LE, no significant results were found in all frequency bands (Figure 1A-right). Next, we compared GE and LE while participants performed the MA task. Notably, the result showed that GE of all frequency bands in the HC group was significantly higher than that in the PD group (Figure 1B-left) while there was no significance in LE (Figure 1B-right).
Next, we tested the interaction effect of the task (RS, MA). In other words, we examined the changes in GE and LE of the two HC and PD groups in response to the changes in the task conditions, from RS to MA. It was clear that in the alpha band, HC's GE was significantly increased while PD's GE was significantly diminished (p < 0.01) (Figure 1C-Alpha). In addition, PD's GE in the beta band was significantly decreased while there were no changes in HC's GE (Figure 1C-Beta). With regards to LE, both HC and PD showed no effect in all bands (Figure 1D).
Next, we tested the interaction effect of the task (RS, MA). In other words, we examined the changes in GE and LE of the two HC and PD groups in response to the changes in the task conditions, from RS to MA. It was clear that in the alpha band, HC's GE was significantly increased while PD's GE was significantly diminished (p < 0.01) (Figure 1C-Alpha). In addition, PD's GE in the beta band was significantly decreased while there were no changes in HC's GE (Figure 1C-Beta). With regards to LE, both HC and PD showed no effect in all bands (Figure 1D).
·Figure 1. Graph theoretical analyses of EEG from patients with panic disorder (PD) and healthy control (HC) in the two different task conditions (Resting state;RS, and Mental arithmetic task; MA).
Conclusions:
These results suggest that the PD group showed reduced GE in brain networks in all frequency bands compared to the HC group in general. This gives an indication that overall information processing efficiency is likely to be decreased in PD as the connectivity of brain networks decreases. We will further examine the role of each band in the connectivity of PD.
Disorders of the Nervous System:
Psychiatric (eg. Depression, Anxiety, Schizophrenia) 1
Learning and Memory:
Working Memory
Modeling and Analysis Methods:
Connectivity (eg. functional, effective, structural) 2
Novel Imaging Acquisition Methods:
EEG
Keywords:
Electroencephaolography (EEG)
Other - panic disorder;Graph theory;Connectivity;Information processing effiency
1|2Indicates the priority used for review
Provide references using author date format
Howard, M. W. (2003), 'Gamma oscillations correlate with working memory load in humans', Cerebral cortex, 13(12), pp. 1369-1374.
Imperatori, C. (2019), 'Default mode network alterations in individuals with high-trait-anxiety: an EEG functional connectivity study', Journal of Affective Disorders, 246, pp. 611-618.
Ismail, L. E. (2020), 'A graph theory-based modeling of functional brain connectivity based on eeg: A systematic review in the context of neuroergonomics', IEEE Access, 8, 155103-155135.
Klimesch, W. (2012), 'Alpha-band oscillations, attention, and controlled access to stored information', Trends in cognitive sciences, 16(12), pp. 606-617.
Lachaux, J. P. (1999), 'Measuring phase synchrony in brain signals', Human brain mapping, 8(4), pp. 194-208.
Latora, V. (2001), 'Efficient behavior of small-world networks. Physical review letters', 87(19), 198701.
Spitzer, B., & Haegens, S. (2017), 'Beyond the status quo: a role for beta oscillations in endogenous content (re) activation', eneuro, 4(4).