Dae Hyeon Kim RESEARCH 한국어

Longitudinal mental and metabolic state modeling

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

Role
Research contributor
Contribution
Constructing activity, sleep, HRV and skin-temperature features. Studying heterogeneous graphs and lagged information flow.
Period & affiliation
2026.04 ~ Present IITP

01 / RESEARCH DETAILLongitudinal observation and evaluation design

Challenge
Identify wearable signals that precede changes in mental and metabolic states despite irregular sampling and missing observations.
Research scope
Represent activity, sleep, HRV and skin temperature as a heterogeneous graph and analyze lagged relationships.
Contribution
Contributing to feature construction, graph modeling and longitudinal evaluation design.
Evaluation conditions and scopeModel and longitudinal evaluation protocol in development.