BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Deformable image registration of abdominal CT images using deep le
 arning: challenges and opportunities - Maureen van Eijnatten
DTSTART:20191004T130000Z
DTEND:20191004T140000Z
UID:TALK129484@talks.cam.ac.uk
CONTACT:J.W.Stevens
DESCRIPTION:In the past few years\, a variety of different methods have be
 en proposed to use deep learning for deformable medical image registration
 . However\, certain challenges need to be overcome before such methods can
  be readily used in clinical settings. For example\, how do these methods 
 deal with small amounts of training data or challenging registrations? Whi
 ch characteristics of displacement fields are prohibitive when using deep 
 learning? And which image similarity metrics and evaluation methods are mo
 st informative when quantifying the performance of such methods? This lect
 ure will demonstrate how such questions may be answered using a longitudin
 al dataset of abdominal CT images. In addition\, a novel training strategy
  is proposed based on transfer learning and displacement field simulations
 .
LOCATION:MR11\, Centre for Mathematical Sciences\, Wilberforce Road\, Camb
 ridge
END:VEVENT
END:VCALENDAR