BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20250513T165422EDT-7482uHjtcG@132.216.98.100
DTSTAMP:20250513T205422Z
DESCRIPTION:Junwei Lu\, PhD\n\nAssistant Professor of Biostatistics\n	Depart
 ment of Biostatistics |\n	Harvard T.H. Chan School of Public Health\n\nAbst
 ract\n\nDue to the increasing adoption of electronic healthrecords (EHR)\,
  large scale EHRshave become another rich data source for translational cl
 inical research. We propose to infer the conditionaldependency structure a
 mong EHR features via a latent graphical block model (LGBM).The LGBM has a
  two layer structure with the first providing semantic embedding vector(SE
 V) representation for the EHR features and the second overlaying a graphic
 al blockmodel on the latent SEVs. The block structures on the graphical mo
 del also allows us tocluster synonymous features in EHR. We propose to lea
 rn the LGBM efficiently\, in bothstatistical and computational sense\, bas
 ed on the empirical point mutual informationmatrix. We establish the stati
 stical rates of the proposed estimators and show the perfectrecovery of th
 e block structure. Numerical results from simulation studies and real EHRd
 ata analyses suggest that the proposed LGBM estimator performs well in fin
 ite sample.\n\nSpeaker bio\n\nJunwei Lu is an Assistant Professor of Biost
 atistics\, Department of Biostatistics\, Harvard T.H. Chan School of Publi
 c Health. His research focuses on the intersection of statistical machine 
 learning and clinical studies\, revealing scientific associations among th
 e clinical treatment strategies and patient phenotyping\, especially focus
 ing on precision medicine leveraging real-world clinical data such as elec
 tronic health records data for risk prediction and clinical optimization.
 \n
DTSTART:20240403T193000Z
DTEND:20240403T203000Z
SUMMARY:Knowledge Graph Embedding with ElectronicHealth Records Data
URL:/epi-biostat-occh/channels/event/knowledge-graph-e
 mbedding-electronichealth-records-data-355691
END:VEVENT
END:VCALENDAR