Projects

Clinical biosignals

Alcohol craving physiomarkers

Individual physiological responses were analyzed to identify candidate markers of alcohol craving.

Technical work-package lead

Graph learning

EEG graph learning and compact decoding

Graph contrastive learning and compact temporal models were developed for EEG emotion recognition and motor-imagery decoding.

EEG research lead

Multimodal learning

Multimodal emotion recognition

Built an older-adult speech dataset and evaluated graph-based fusion of speech and text for emotion recognition.

Technical work-package lead

Graph learning

Multimodal time-series fusion

Fused signal-specific similarity networks to learn multimodal representations with limited labels.

Technical work-package lead

05

Ongoing research

Longitudinal mental and metabolic state modeling

Researching lagged relationships between mental and metabolic state changes using longitudinal wearable observations.

Research contributor

Ongoing research

Dual-stream interaction modeling

Researching a dual-stream architecture that connects continuous physiological encoders with dialogue models.

Research contributor and model co-developer

Domain adaptation

Multimodal data and domain adaptation

Evaluated multimodal fusion and adversarial domain adaptation under subject and session changes.

Research contributor; co-first author of MDA-GCL

Applied research

Contactless vital estimation and deployment

Adapted facial pulse-estimation models to Jetson hardware and supported kiosk deployment.

Algorithm co-development and inference optimization

KIST internship

Focused-ultrasound registration and control

Registered imaging and tracking coordinates and implemented position control for focused-ultrasound field measurement.

3D Slicer module and device-control developer

Undergraduate internship

Blood pressure estimation from two-channel BCG

Estimated blood pressure from two chair BCG channels using a CNN, BiGRU and temporal attention.

Co-developer of blood-pressure estimation and real-time processing

AI Champion 2026

NeuroSync

Combined Korean LLMs with evidence verification to turn intake conversations into structured clinician handoff records.

Method design and core implementation

AI Champion 2026

NeuroTruth

Connected wearable craving estimates and rule-based risk decisions to app notifications and conversational support.

Technical lead and project manager

Research internship

Multiscale entropy for short EEG

Used moving averages to retain samples when estimating multiscale entropy from short EEG segments.

Method comparison and co-first authorship