Underlying Mechanisms of Aging and Neurological Disorders
Underlying Mechanisms of Aging and Neurological Disorders
We investigate how pathological changes in the brain affect neural activity during memory processing and contribute to memory decline. Based on evidence of early spatiotemporal memory deficits, we aim to develop a novel approach for the early prediction of Alzheimer’s disease.
Brain age can differ from chronological age, reflecting individual differences in brain aging. We use AI-based models to estimate brain age from fMRI and EEG signals and aim to predict individual trajectories of brain aging and cognitive decline using longitudinal data.
Simultaneous EEG-fMRI combines the high temporal resolution of EEG with the high spatial resolution of fMRI to understand how the brain works.
Functional near-infrared spectroscopy (fNIRS) enables extended monitoring of brain activity with fewer movement constraints, supporting research in naturalistic settings.
Physiological signals, including PPG, skin conductance, and eye gaze, provide complementary insights into cognitive and emotional states, supporting research on anxiety and depression.
Digital Healthcare Solutions for Mental Health
(In collaboration with Dr. Sang Ah Lee at Seoul National University)
We develop digital app-based tasks to assess cognitive functions, monitor individual mental health, and predict the onset of depression and anxiety disorders.
We develop gamified cognitive training to improve attention, emotion regulation, and social functioning in everyday life, with a focus on keeping users engaged and motivated to train regularly.
Machine learning or AI-driven models analyze large-scale datasets to predict individual mental health profiles, including anxiety, depressive symptoms, self-esteem, and ADHD symptoms.
Portable EEG devices facilitate brain signal recording in everyday settings to support brain health monitoring.
Adaptive digital therapy and personalization
This framework iteratively evaluates stimulation parameters and updates a Gaussian process model to optimize neuromodulation for each individual.
Cognitive training and neuromodulation are tailored to individual neurocognitive profiles characterized by cognitive performance and neural activity.
AI-based dimensionality reduction reveals latent neural dynamics, characterizing neural state trajectories during memory encoding and retrieval.