BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Generative model-based super resolution and quality control for ca
 rdiac segmentation - Shuo Wang (Digital Medicine Research Centre\, Fudan U
 niversity)
DTSTART:20210629T120000Z
DTEND:20210629T130000Z
UID:TALK161233@talks.cam.ac.uk
CONTACT:J.W.Stevens
DESCRIPTION:In cardiac imaging\, a high-resolution geometric representatio
 n of the heart is desired for accurate assessment of its anatomical struct
 ure and function. This is not easily available due to the limit of acquisi
 tion duration and respiratory/cardiac motion in clinical practice. Stacks 
 of multi-slice 2D images are usually acquired in clinical routine and segm
 entation of these images provides a low-resolution representation of cardi
 ac anatomy\, which may contain artefacts caused by motion. Here we propose
  a novel latent optimisation framework that jointly performs motion correc
 tion and super resolution for cardiac image segmentations based on generat
 ive learning. Moreover\, quality control of the automatic segmentation res
 ults is realised via the proposed framework.\n\nZoom link:\nhttps://maths-
 cam-ac-uk.zoom.us/j/92575403744?pwd=RHhqWC9wcUVWQi9xSzc1UE9BVGk3Zz09\n\nMe
 eting ID: 925 7540 3744\nPasscode: 974971\n
LOCATION:Virtual (see abstract for Zoom link)
END:VEVENT
END:VCALENDAR