BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20210916T132452Z
LOCATION:Henry Dunant
DTSTART;TZID=Europe/Stockholm:20210707T150000
DTEND;TZID=Europe/Stockholm:20210707T153000
UID:submissions.pasc-conference.org_PASC21_sess152_msa260@linklings.com
SUMMARY:Health Data Science at Scale
DESCRIPTION:Minisymposium\n\nHealth Data Science at Scale\n\nGounley\n\nIn
 formation extraction from clinical text documents is an important tool for
  cancer surveillance and research, with phenotyping used to stratify patie
 nts into cohorts based on tumor biology, responsiveness to treatment, and 
 many other factors. To facilitate this work, natural language processing t
 echniques using deep learning have been developed to automatically extract
  information from unstructured clinical text documents such as cancer path
 ology reports. Recently developed Transformer-based deep learning architec
 tures like BERT offer significant potential for improving the accuracy and
  robustness of information extraction from clinical text. At the same time
 , training of BERT models is much more computationally intensive than alte
 rnative deep learning approaches. In this talk, we consider the effectiven
 ess of BERT models for clinical text classification and opportunities to i
 mprove performance on HPC resources.\n\nDomain: CS and Math, Life Sciences
END:VEVENT
END:VCALENDAR