PSYCHIATRIC EPIDEMIOLOGY LAB

Director: P. Daniel Lin, MD, PhD

The Psychiatric Epidemiology & Multimodal Data Lab (PsychEpi Lab) investigates the biological, behavioral, and environmental determinants of suicide risk, neurodevelopmental conditions, and mental-health outcomes across the lifespan. Our work integrates large-scale population data with fine-grained digital and neurobiological measures to build scalable, evidence-based tools for prediction, prevention, and clinical decision support.

Our Multimodal Approach

We combine complementary data sources to capture the complexity of psychiatric phenomena, including:

• Electronic Health Records (EHRs)
  Longitudinal clinical data from multi-site health systems are used to model risk trajectories, treatment response, and health-care utilization patterns.

• Genomic & Bioinformatic Data
  GWAS summary statistics, polygenic risk scores, Neanderthal-derived variants, and transcriptomic pathways are analyzed to uncover genetic architectures and gene–environment interactions relevant to suicidality and neurodevelopment.

• Digital Biomarkers & Behavioral Data
  We incorporate eye-tracking, fNIRS neuroimaging, smartphone usage patterns, and real-time ecological momentary assessments (EMA) to understand emotion regulation, attention, and impulsivity in naturalistic settings.

• Social & Environmental Determinants
  We leverage community-level datasets, social determinants of health indicators, and public data repositories to contextualize individual-level psychiatric risk within broader environmental systems.

Research Themes

• Suicide Risk & Resilience
  Identifying protective pathways—including psychiatric, genetic, and neurocognitive mechanisms—that enable resilience despite elevated risk.

• Digital Phenotyping & Precision Mental Health
  Using multimodal behavioral and neurobiological signals to develop personalized risk-stratification tools and adaptive clinical interventions.

• Neurodevelopment & Youth Mental Health
  Understanding how visual attention, impulsivity, and emotion arousal contribute to suicidal ideation among adolescents with ADHD and autism.

• Data-Driven Clinical Decision Support
  Applying machine learning to EHR and biosocial data to inform treatment intensity, resource allocation, and crisis-care pathways.