Sequence based clustering of functional brain networks and its association with negative emotions
Poster No:
1794
Submission Type:
Abstract Submission
Authors:
Dong Yun Lee1, Seul-Gi Lee2, Rae Woong Park2, Bumhee Park2
Institutions:
1Department of Biomedical Informatics, Ajou University School of Medicine, Suwon, gyeonggi-do, 2Department of Biomedical Informatics, Ajou University School of Medicine, Suwon, gyeonggi-do
First Author:
Dong Yun Lee
Department of Biomedical Informatics, Ajou University School of Medicine
Suwon, gyeonggi-do
Department of Biomedical Informatics, Ajou University School of Medicine
Suwon, gyeonggi-do
Co-Author(s):
Seul-Gi Lee
Department of Biomedical Informatics, Ajou University School of Medicine
Suwon, gyeonggi-do
Department of Biomedical Informatics, Ajou University School of Medicine
Suwon, gyeonggi-do
Rae Woong Park
Department of Biomedical Informatics, Ajou University School of Medicine
Suwon, gyeonggi-do
Department of Biomedical Informatics, Ajou University School of Medicine
Suwon, gyeonggi-do
Bumhee Park
Department of Biomedical Informatics, Ajou University School of Medicine
Suwon, gyeonggi-do
Department of Biomedical Informatics, Ajou University School of Medicine
Suwon, gyeonggi-do
Introduction:
Resting-state functional MRI (rs-fMRI) captures patterns of spontaneous brain activity that can reveal clues about the connectome of the human brain. While many rs-fMRI studies have focused on static measures of functional connectivity, examining the temporal dynamics of rs-fMRI data provides a complementary view of the functional connectome, helping to identify changes in disease, development, and aging [1]. Meanwhile, Studies using rs-fMRI have reported relationships between negative emotions and brain connectivity [2]. This study aimed to propose a sequence-based clustering of rs-fMRI and investigate the relation between clustering and negative emotions.
Methods:
For our study, we used the resting state fMRI (rsfMRI) and scales for the US National Institutes of Health (NIH) Toolbox Emotion Battery from 1080 healthy subjects, which were provided by the Human Connectome Project (HCP) S1200 release dataset. The rsfMRI data were canonically preprocessed through realignment, co-registration, normalization, smoothing, regressing out motion related confounding factors, despiking, and band-pass filtering (0.01 ~ 0.1 Hz). Brain regions were then defined with Brainnetome atlas [3] and Schaefer 200 [4]. Among these regions, individual inter-regional functional connectivity was calculated using a Pearson correlation coefficient and Fisher's r-to-z transformation. We performed an independent component analysis (ICA) on those brain networks and estimated the time series. Using time series, we did sequence-based clustering using Sequence Graph Transform (SGT) and spectral clustering on acquired brain network sequences [5]. Finally, emotion scales were reduced to meaningful combinations with the principal components analysis and associated with the sequence-based clusters.
Results:
From the ICA, we identified several components such as DMN and salience network. Based on this, we obtained a sequence of independent components for each person and further found three clusters (Figure 1). Two principal component scores for negative emotions were significantly different according to clusters (Figure 2, p <0.05).
·Figure 1
·Figure 2
Conclusions:
We found that the sequential clusters of rs-fMRI were related to negative emotions. Our findings suggest that the temporal dynamics of rs-fMRI data may be a more useful option for investigating clinical symptoms. Replication of our results from using the temporal approach are needed by other groups.
Emotion, Motivation and Social Neuroscience:
Emotion and Motivation Other 2
Modeling and Analysis Methods:
fMRI Connectivity and Network Modeling 1
Neuroanatomy, Physiology, Metabolism and Neurotransmission:
Anatomy and Functional Systems
Neuroinformatics and Data Sharing:
Workflows
Novel Imaging Acquisition Methods:
BOLD fMRI
Keywords:
ADULTS
Design and Analysis
Emotions
FUNCTIONAL MRI
Statistical Methods
Workflows
1|2Indicates the priority used for review
Provide references using author date format
[2] Canario E et al. (2021) A review of resting-state fMRI and its use to examine psychiatric disorders. Psychoradiology. 1(1):42-53.
[3] Fan, L., et al. (2016). The human brainnetome atlas: a new brain atlas based on connectional architecture. Cerebral cortex, 26(8), 3508-3526.
[4] Kong, R., et al. (2021). Individual-specific areal-level parcellations improve functional connectivity prediction of behavior. Cerebral Cortex, 31(10), 4477-4500.
[5] Ranjan C., et al. (2022) Sequence graph transform (SGT): a feature embedding function for sequence data mining. Data Mining and Knowledge Discovery. 36(2):668-708.