BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Classifying Stroke Using Electrical Impedance Tomography - Samulti
  Siltanen (University of Helsinki)
DTSTART:20191010T150000Z
DTEND:20191010T160000Z
UID:TALK131269@talks.cam.ac.uk
CONTACT:Carola-Bibiane Schoenlieb
DESCRIPTION:Stroke is a leading cause of death all around the world. There
  are two main types of stroke:\nischemic (blood clot preventing blood flow
  to a part of the brain) and hemorrhagic (bleeding in\nthe brain). The sym
 ptoms are the same\, but treatments very different. A portable "stroke\ncl
 assifier" would be a life-saving equipment to have in ambulances\, but so 
 far it does not exist.\nElectrical Impedance Tomography (EIT) is a promisi
 ng and harmless imaging method for stroke\nclassification. In EIT one atte
 mpts to recover the electric conductivity inside a domain from\nelectric b
 oundary measurements. This is a nonlinear and ill-posed inverse problem. T
 he so-called\nComplex Geometric Optics (CGO) solutions have proven to be a
  useful computational tool for\nreconstruction tasks in EIT. A new propert
 y of CGO solutions is presented\, showing that a one-dimensional Fourier t
 ransform in the spectral variable provides a connection to parallel-beam X
 ray tomography of the conductivity. One of the consequences of this “non
 linear Fourier slice theorem” is a novel capability to recover inclusion
 s within inclusions in EIT. In practical imaging\, measurement noise cause
 s strong blurring in the recovered profile functions. However\, machine le
 arning algorithms can be combined with the nonlinear PDE techniques in a f
 ruitful way. As an\nexample\, simulated strokes are classified into hemorr
 hagic and ischemic using EIT measurements.
LOCATION:MR 13
END:VEVENT
END:VCALENDAR