Byung-Hoon Kim

Graph neural networks (GNN) rely on graph operations that include neural network training for various graph related tasks. Recently, several attempts have been made to apply the GNNs to functional magnetic resonance image (fMRI) data. Despite recent progresses, a common limitation is its difficulty to explain the classification results in a neuroscientifically explainable way. Here, we develop a framework for analyzing the fMRI data using the Graph Isomorphism Network (GIN), which was recently p

Functional connectivity (FC) between regions of the brain can be assessed by the degree of temporal correlation measured with functional neuroimaging modalities. Based on the fact that these connectivities build a network, graph-based approaches for analyzing the brain connectome have provided insights into the functions of the human brain. The development of graph neural networks (GNNs) capable of learning representation from graph structured data has led to increased interest in learning the g

O6-methylguanine-DNA methyl transferase (MGMT) methylation prediction models were developed using only small datasets without proper external validation and achieved good diagnostic performance, which seems to indicate a promising future for radiogenomics. However, the diagnostic performance was not reproducible for numerous research teams when using a larger dataset in the RSNA-MICCAI Brain Tumor Radiogenomic Classification 2021 challenge. To our knowledge, there has been no study regarding the

Social anxiety is a symptom widely prevalent among young adults, and when present in excess, can lead to maladaptive patterns of social behavior. Recent approaches that incorporate brain functional radiomic features and machine learning have shown potential for predicting certain phenotypes or disorders from functional magnetic resonance images. In this study, we aimed to predict the level of social anxiety in young adult participants by training machine learning models with resting-state brain

Background Mental health issues, including panic disorder (PD), are prevalent and often co-occur with anxiety and bipolar disorders. While panic disorder is characterized by unexpected panic attacks, and its treatment often involves antidepressants, there is a 20–40% risk of inducing mania (antidepressant-induced mania) during treatment, making it crucial to understand mania risk factors. However, research on clinical and neurological characteristics of patients with anxiety disorders who develo

Social anxiety disorder (SAD) is a mental disorder characterized by excessive anxiety in social situations. This study aimed to examine the alteration of resting-state functional connectivity in SAD patients related to the virtual reality-based self-training (VRS) which enables exposure to social situations in a controlled environment. Fifty-two SAD patients were randomly assigned to the experimental group who received the VRS, or the control group who did not. Self-report questionnaires and res

본 연구는 진지한 여가로서 모터사이클 참여자들의 진지한 여가 경험을 분석하는데 그 목적이 있다. 연구자는 본 연구의 목적에 부합할 수 있도록 현상에 대한 의미 파악이 가능한 현상학적 연구방법을 연구 틀로 연구의 문제들을 해결하고자 하였다. 연구 참여자들은 새로움에 대한 호기심과 타인에 의한 권유로 모터사이클을 시작하여 긍정적인 체험과 부정적인 체험을 동시에 경험하고 모터사이클을 통해 상쾌함이나 스스로의 실력에 대한 자부심, 성공 경험에 오는 성취, 열정을 체험한다. 일상적 여과와는 상대적인 의미를 가진 진지한 여가 활동을 하는 연구참여자들이 모터사이클을 타는 시즌에 다른 사회적 관계가 단절 되는 것을 발견할 수 있으며 이를 통해 갈등이 형성되고 상황적, 금전적, 입지적, 시간적 제약을 체험한다. 이상과 같이 연구 참여자들은 긍정적인 체험을 하는 동시에 부정적 체험을 하는데 이는 포기하거나 중단하는 계기를 형성하지 않고 오히려 적극적인 노력과 투자를 통해 자신만의 특이한 행동 양식의

Introduction Recurrences and diagnostic instability of panic disorder (PD) are common and have a negative effect on its long-term course. Developing a novel assessment tool for anxiety that can be used in a multimodal approach may improve these problems in panic disorder patients. This study assessed the feasibility of virtual reality-based assessment in panic disorder (VRA-PD). Methods Twenty-five patients with PD (ANX group) and 28 healthy adults (CON group) participated in the study. VRA-PD c

Graph neural networks (GNN) rely on graph operations that include neural network training for various graph related tasks. Recently, several attempts have been made to apply the GNNs to functional magnetic resonance image (fMRI) data. Despite recent progresses, a common limitation is its difficulty to explain the classification results in a neuroscientifically explainable way. Here, we develop a framework for analyzing the fMRI data using the Graph Isomorphism Network (GIN), which was recently p

Background North Korean defectors (NKDs) have often been exposed to traumatic events. However, there have been few studies of neural alterations in NKDs with post-traumatic stress disorder (PTSD) and complex PTSD (cPTSD). Aims To investigate neural alterations in NKDs with PTSD and cPTSD, with a specific focus on alterations in resting-state functional connectivity networks, including the default mode network (DMN). Method Resting-state functional connectivity was assessed using brain functional

Anhedonia is one of the major negative symptoms in schizophrenia and defined as the loss of hedonic experience to various stimuli in real life. Although structural magnetic resonance imaging has provided a deeper understanding of anhedonia-related abnormalities in schizophrenia, network analysis of the grey matter focusing on this symptom is lacking. In this study, single-subject grey matter networks were constructed in 123 patients with schizophrenia and 160 healthy controls. The small-world pr

Graph Transformers have recently been successful in various graph representation learning tasks, providing a number of advantages over message-passing Graph Neural Networks. Utilizing Graph Transformers for learning the representation of the brain functional connectivity network is also gaining interest. However, studies to date have underlooked the temporal dynamics of functional connectivity, which can reflect important markers of brain function. Here, we propose a method for learning the repr