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:20230627T170000
DTEND;TZID=Europe/Stockholm:20230627T173000
UID:submissions.pasc-conference.org_PASC23_sess174_msa133@linklings.com
SUMMARY:Scientific Machine Learning for Cardiac Electrophysiology Applicat
 ions
DESCRIPTION:Minisymposium\n\nSimone Pezzuto (University of Trento, Univers
 itą della Svizzera italiana); Francisco Sahli Costabal (Pontificia Univers
 idad Católica de Chile); and Paris Perdikaris (University of Pennsylvania)
 \n\nCardiac modeling for precision cardiology is an emerging technology in
  clinical practice. Thanks to the sophistication of state-of-the-art elect
 rophysiology models, it is possible to tailor treatments to patient charac
 teristics, thus improving the therapeutic outcome. Patient-specific modeli
 ng requires a deep integration of clinical data into existing models. This
  aspect is, however, not straightforward. Cardiac models are computational
 ly expensive, with several patient-specific parameters of difficult identi
 fication. Clinical data is scarce, multi-modal, and sparse in space-time. 
 Thus, neither purely data-driven nor model-driven approaches are optimal i
 n the digital twinning process. In this talk, we will solve two clinically
 -relevant problems using a physics-informed strategy. The first problem co
 nsists in recovering the conductivity tensor in the heart, starting from s
 parse electric recordings collected by clinicians. We propose FiberNet, a 
 physics-informed neural network method for solving the corresponding inver
 se problem. We weakly impose the physiological electric propagation with t
 he anisotropic eikonal model. The second application concerns atrial fibri
 llation inducibility in a complex anatomical model of human atria. We prop
 ose a multi-fidelity classifier that learns the inducibility map on a mani
 fold. Finally, we generalize the classifier so that it does not require an
 y new simulation when the anatomy changes.\n\nDomain: Life Sciences\n\nSes
 sion Chair: Georgios Kissas (University of Pennsylvania)
END:VEVENT
END:VCALENDAR