BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Uncertainty quantification and sampling optimization for deep lear
 ning-based medical imaging
DTSTART:20190829T100000
DTEND:20190829T120000
DTSTAMP:20260407T164121Z
UID:19c283dda6f2fe96db2fc1a4cb9850d44224fdfdd13971332e02778a
CATEGORIES:Conferences - Seminars
DESCRIPTION:Thomas Sanchez\nEDIC candidacy exam\nExam president: Dr. Fran
 çois Fleuret\nThesis advisor: Prof. Volkan Cevher\nCo-examiner: Prof. Dim
 itri Van De Ville\n\nAbstract\nFast\, reliable and accurate solutions to i
 nverse problems have always been of prime importance to the medical imagin
 g community. In this report\, we discuss recent advances in three compleme
 ntary elements of any imaging pipeline. First\, we dive into one of the ea
 rliest deep learning-based reconstruction methods for MRI\, which achieved
  unprecedented quality and speed of reconstruction compared to iterative m
 ethods. Secondly\, we discuss the problem of sampling optimization\, where
  one seeks the best sampling pattern for a given reconstruction algorithm.
  However\, the clinical deployment of any of these methods would require a
 ccess to both mean and variance estimates\, in order to prevent critical d
 iagnoses from relying on uncertain features. A solution to this problem is
  presented in the last part of this report\, where we cover an approach th
 at allows tractable sampling from the posterior distribution of a reconstr
 ucted image.\n\nBackground papers\nA Deep Cascade of Convolutional Neural 
 Networks for Dynamic MR Image Reconstruction\, by Schlemper\, J. et al.\n
 Learning-Based Compressive MRI\, by Gözcü\, B. et al. \nDeep Bayesian 
 Inversion\, by Adler\, J. and Öktem\, O.\n 
LOCATION:DIA 005 https://plan.epfl.ch/?room==DIA%20005
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR