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:20190719T085743Z
LOCATION:HG F 3
DTSTART;TZID=Europe/Stockholm:20190612T133000
DTEND;TZID=Europe/Stockholm:20190612T140000
UID:submissions.pasc-conference.org_PASC19_sess117_msa204@linklings.com
SUMMARY:How Deep Learning on HPC Systems Enables Novel Approaches for Mapp
 ing the Human Brain
DESCRIPTION:Minisymposium\nComputer Science and Applied Mathematics, Physi
 cs, Life Sciences\n\nHow Deep Learning on HPC Systems Enables Novel Approa
 ches for Mapping the Human Brain\n\nDickscheid, Schiffer, Spitzer, Lippert
 , Amunts\n\nThe human brain with its approximately 86 billion nerve cells 
 is one of the most complex systems. It is segregated into areas, whic
 h differ in structure and function. Multi-level atlases are needed to
  provide a spatial reference for different experimental data and mode
 ls at various spatial and temporal scales, and to integrate, exchange
  and compare data. The Human Brain Project (HBP) is building such an 
 atlas, which encompasses a reference brain at the microscopic level. 
 To analyze neurons and to build 3D maps of brain areas at cellular le
 vel, Deep Learning and HPC are being employed. Histological sections 
 are being imaged by high-throughput microscopes, resulting in large i
 mages of 200x100k pixels. In order to perform an automated analysis of&nbs
 p;several thousands of images of histological sections, algorithms for bra
 in mapping have been developed, which process image data at the Petab
 yte scale. Hereby, we rely on multi-GPU distributed memory environmen
 ts provided by the Juelich Supercomputing Center. Workflows include r
 emote visualization and distributed data management between the lab a
 nd HPC center. The combination of high-throughput microscopy with HPC
  based workflows provides a challenging technological use case, which
  opens new perspectives to compute cellular human brain models.
END:VEVENT
END:VCALENDAR