BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//wp-events-plugin.com//7.4.0.1//EN
TZID:Europe/Paris
X-WR-TIMEZONE:Europe/Paris
BEGIN:VEVENT
UID:121@isdm.umontpellier.fr
DTSTART;TZID=Europe/Paris:20250203T170000
DTEND;TZID=Europe/Paris:20250203T170000
DTSTAMP:20260602T130232Z
URL:https://isdm.umontpellier.fr/events/personalizing-treatment-with-causa
 l-inference-and-scalably-evaluating-llms-in-medicine-2/
SUMMARY:Personalizing Treatment with Causal Inference and Scalably Evaluati
 ng LLMs in Medicine
DESCRIPTION:\nFrançois Grolleau\n\nThis talk examines two critical aspects
  of data-driven medicine: personalized treatment strategies using causal i
 nference and the robust evaluation of large language models (LLMs) for cli
 nical applications.\n\nIn the first part\, we present a novel approach to 
 personalized medicine\, applying causal statistical learning to observatio
 nal data to develop individualized treatment rules. We focus on optimizing
  the timing of renal replacement therapy initiation in acute kidney injury
 \, demonstrating: (i) the estimation and validation of an optimal dynamic 
 strategy\, and (ii) a comprehensive framework for evaluating individualize
 d rules using observational data.\n\nIn the second part\, we tackle the ch
 allenge of evaluating LLMs in medicine\, focusing on the generation of hos
 pital course summaries. Current evaluation methods are often either unscal
 able (physician-led) or untrustworthy for clinical settings (LLM-as-a-judg
 e). We propose a rubric-based approach to LLM evaluation that combines the
  scalability of automated methods with the trustworthiness demanded by med
 ical applications\, paving the way for responsible deployment of LLMs in h
 ealthcare.\n\nData-driven medicine\, LLMs\, clinical applications\, Labél
 iséHallesIA
ATTACH;FMTTYPE=image/jpeg:https://isdm.umontpellier.fr/wp-content/uploads/
 2025/02/ml-mpt2028129-TM2f7V.tmp_.jpg
END:VEVENT
BEGIN:VTIMEZONE
TZID:Europe/Paris
X-LIC-LOCATION:Europe/Paris
BEGIN:STANDARD
DTSTART:20241027T020000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
END:STANDARD
END:VTIMEZONE
END:VCALENDAR