Accelerating magnetic induction tomography-based imaging through heterogeneous parallel computing
- Submitting institution
-
Cardiff University / Prifysgol Caerdydd
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 96067594
- Type
- D - Journal article
- DOI
-
10.1002/cpe.5265
- Title of journal
- Concurrency and Computation: Practice and Experience
- Article number
- e5265
- First page
- -
- Volume
- 31
- Issue
- 17
- ISSN
- 1532-0626
- Open access status
- Compliant
- Month of publication
- April
- Year of publication
- 2019
- URL
-
http://dx.doi.org/10.1002/cpe.5265
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
4
- Research group(s)
-
C - Cybersecurity, privacy and human centred computing
- Citation count
- 0
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Magnetic induction tomography (MIT) is a cheap, portable, non-invasive imaging technique that has applications in industry and healthcare, such as non-destructive testing and the treatment of cerebral strokes. Funded from an EPSRC grant (EP/K024078/1: £430k) this paper shows how heterogeneous parallelism, using multiple cores and multiple graphical processing units (GPUs), can reduce the processing time for MIT to clinically relevant timescales. The paper also demonstrates how an existing third-party mathematical software library (deal.II) can be extended to make use of an arbitrary number of GPUs.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -