Lazy Resampling: Fast and information preserving preprocessing for deep learning
Lazy Resampling reduces the loss of information that occurs when running processing pipelines that traditionally have multiple resampling steps and enables researchers to build simpler pipelines by making operations such as rotation and cropping effectively non-destructive. It makes it possible to invert labels back through a pipeline without catastrophic loss of accuracy. A reference implementation for Lazy Resampling can be found at https://github.com/KCL-BMEIS/LazyResampling. Lazy Resampling is being implemented as a core feature in MONAI, an open source python-based deep learning library for medical imaging, with a roadmap for a full integration.
Authors
- Marc Modat (ORCID: https://orcid.org/0000-0002-5277-8530)
- Richard Brown (ORCID: https://orcid.org/0000-0001-6989-9200)
- Daguang Xu (ORCID: https://orcid.org/0000-0002-4621-881X)
- Andrew Feng (ORCID: https://orcid.org/0000-0002-1675-745X)
- Eric Kerfoot (ORCID: https://orcid.org/0000-0002-7255-6417)
- M. Jorge Cardoso (ORCID: https://orcid.org/0000-0003-1284-2558)
- Benjamin Murray (ORCID: https://orcid.org/0000-0002-2413-923X)
- Sébastien Ourselin (ORCID: https://orcid.org/0000-0002-5694-5340)
- Pengcheng Ma
Institutions
- King's College London (GB)
- Nvidia (United States) (US)
Publication Details
- Journal
- Computer Methods and Programs in Biomedicine
- Published
- 2024-09-19
- DOI
- https://doi.org/10.1016/j.cmpb.2024.108422
- Citations
- 2
- Primary Topic
- Advanced Neural Network Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.50
Funders
- Wellcome Trust
- Engineering and Physical Sciences Research Council
- Centre For Medical Engineering, King’s College London