Overcoming Measurement Inconsistency In Deep Learning For Linear Inverse Problems: Applications In Medical Imaging

Marija Vella, João F. C. Mota

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)
3 Downloads (Pure)

Abstract

The remarkable performance of deep neural networks (DNNs) currently makes them the method of choice for solving linear inverse problems. They have been applied to super-resolve and restore images, as well as to reconstruct MR and CT images. In these applications, DNNs invert a forward operator by finding, via training data, a map between the measurements and the input images. It is then expected that the map is still valid for the test data. This framework, however, introduces measurement inconsistency during testing. We show that such inconsistency, which can be critical in domains like medical imaging or defense, is intimately related to the generalization error. We then propose a framework that post-processes the output of DNNs with an optimization algorithm that enforces measurement consistency. Experiments on MR images show that enforcing measurement consistency via our method can lead to large gains in reconstruction performance.
Original languageEnglish
Title of host publication2021 International Conference on Acoustics, Speech, and Signal Processing
PublisherIEEE
Pages8113-8117
Number of pages5
ISBN (Electronic)9781728176055
DOIs
Publication statusPublished - 13 May 2021
Event46th International Conference on Acoustics, Speech, and Signal Processing 2021 -
Duration: 6 Jun 202111 Jun 2021

Conference

Conference46th International Conference on Acoustics, Speech, and Signal Processing 2021
Abbreviated titleICASSP 2021
Period6/06/2111/06/21

Keywords

  • Linear inverse problems
  • Medical imaging
  • Neural networks
  • Optimization
  • Total variation

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Overcoming Measurement Inconsistency In Deep Learning For Linear Inverse Problems: Applications In Medical Imaging'. Together they form a unique fingerprint.

Cite this