BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Data-driven enhancement of electrical impedance tomography using s
 egmentation flow - Samuli Siltanen (University of Helsinki)
DTSTART:20131202T160000Z
DTEND:20131202T170000Z
UID:TALK49159@talks.cam.ac.uk
CONTACT:Carola-Bibiane Schoenlieb
DESCRIPTION:Electrical Impedance Tomography (EIT) is a non-invasive\, inex
 pensive\, and portable medical imaging modality where the patient is probe
 d with electric currents fed through electrodes positioned on the skin. Th
 e resulting voltages at the electrodes are measured\, and the goal is to r
 ecover the internal electric conductivity of the body. The reconstruction 
 task is a highly ill-posed nonlinear inverse problem and requires the use 
 of regularized solution methods. In the so-called D-bar method\, one uses 
 a nonlinear low-pass filter to provide regularization. However\, this resu
 lts in a blurry reconstruction that obscures crisp boundaries between tiss
 ues. We propose sharpening the EIT image using the diffusive image segment
 ation of Ambrosio and Tortorelli\, controlled by the EIT data vie the so-c
 alled CGO sinogram. 
LOCATION:MR 4\, Centre for Mathematical Sciences
END:VEVENT
END:VCALENDAR