Team members

My research interest is using electrophysiological signals for the screening of depression and suicide risks. Depression and suicidal tendencies are often associated with dysfunctions in the autonomic nervous system and abnormalities in brain network activity. Electrophysiological signals, including heart rate variability (HRV), electroencephalogram (EEG) features, and galvanic skin response (GSR), provide insights into the regulatory states of the sympathetic and parasympathetic nervous systems as well as emotion-related neural activities, providing objective biomarkers for the early detection of depression and suicide. In my work, I collect multimodal electrophysiological data—such as EEG, electrocardiogram (ECG), electromyogram (EMG), and GSR—and apply frequency-domain, time-domain, and nonlinear analytical methods to investigate the correlation between signal variations and depression/suicide risks. Furthermore, deep learning and explainable artificial intelligence techniques can be employed to extract key features and identify recurring patterns in the data, presenting a potential research direction for constructing effective detection models. The overarching goal of this research is to develop a real-time, non-invasive, and highly sensitive screening tool that supports the early detection and personalized intervention of depression and suicide. During my doctoral studies, my primary research focused on pattern recognition for intelligent agents, including a project aimed at developing goal recognition algorithms to analyze EMG signals and identify human arm motion intentions. I published several papers as the first/corresponding author in leading journals such as Artificial Intelligence and Engineering Applications of Artificial Intelligence.