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:20230831T095746Z
LOCATION:Sanada I
DTSTART;TZID=Europe/Stockholm:20230627T113000
DTEND;TZID=Europe/Stockholm:20230627T120000
UID:submissions.pasc-conference.org_PASC23_sess173_msa298@linklings.com
SUMMARY:Leveraging Graph Neural Networks for Efficient Reduced-Order Blood
  Flow Simulations
DESCRIPTION:Minisymposium\n\nLuca Pegolotti, Martin Pfaller, and Natalia R
 ubio (Stanford University); Ke Ding and Rita Brugarolas (Intel Corporation
 ); and Eric Darve and Alison Marsden (Stanford University)\n\nRecently, si
 mulations of blood flow have shown great promise in revolutionizing cardio
 vascular disease research and treatment. Reduced-order models, specificall
 y zero- and one-dimensional ones, can approximate blood dynamics more effi
 ciently than detailed three-dimensional simulations. These models prove he
 lpful when computational resources or time are constrained or in scenarios
  requiring numerous queries, such as uncertainty quantification. However, 
 their accuracy can falter in complex geometries featuring many junctions o
 r pathological conditions like stenoses or aneurysms. Data-driven reduced-
 order models utilize previous simulation data to address the limitations o
 f conventional physics-based models. Our presentation delves into a one-di
 mensional reduced-order model utilizing MeshGraphNet, a graph neural netwo
 rk architecture designed for simulations on unstructured grids. The graph 
 neural network is trained on various three-dimensional simulation data res
 tricted to the nodes on the geometry centerline. Thanks to the versatility
  of graph neural networks, a single trained architecture can effectively a
 pproximate blood dynamics in a range of topologies and geometries. Our res
 ults show that, given sufficient training data, our algorithm outperforms 
 physics-based one-dimensional models in accuracy while still offering dram
 atic speed enhancements compared to three-dimensional simulations. In this
  presentation, we also cover the limitations of the current methodology an
 d future perspectives.\n\nDomain: Life Sciences\n\nSession Chair: Georgios
  Kissas (University of Pennsylvania)
END:VEVENT
END:VCALENDAR