BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Machine Learned Priors for Nonsmooth Conductivities in D-bar Recon
 structions of 2D EIT Data - Melody Alsaker (Gonzaga University)
DTSTART:20230328T135000Z
DTEND:20230328T144000Z
UID:TALK198226@talks.cam.ac.uk
DESCRIPTION:Recent developments in D-bar reconstruction methods for 2D EIT
  have established a methodology for the inclusion of spatial priors\, whic
 h have been shown to provide improved quality and stability of images. In 
 the context of medical imaging\, these techniques begin with prior estimat
 es of organ boundaries within the plane of the electrodes\, to which optim
 ized conductivity guesses are assigned. In previous works\, the methodolog
 y for approximating organ boundaries has involved manually extracting boun
 daries from prior medical scans\, which may not be readily available in pr
 actice. This protocol is also highly labor intensive\, and has the potenti
 al to introduce human bias. Furthermore\, in previous works\, some of the 
 sharpness provided by the introduction of priors was lost due to a mathema
 tical need for smoothing of the conductivity distribution. In this present
 ation\, we address these problems via (1) a method for the automated selec
 tion of boundaries via machine learning techniques\, and (2) use of an alt
 ernative mathematical formulation which eliminates the need for smoothing.
  We present results from numerically simulated thoracic phantoms on circul
 ar domains.
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR