Analysis of tuberculosis severity levels from CT pulmonary images based on enhanced residual deep learning architecture
- Submitting institution
-
Middlesex University
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 787
- Type
- D - Journal article
- DOI
-
10.1016/j.neucom.2018.12.086
- Title of journal
- Neurocomputing
- Article number
- -
- First page
- 233
- Volume
- 392
- Issue
- -
- ISSN
- 0925-2312
- Open access status
- Compliant
- Month of publication
- April
- Year of publication
- 2019
- URL
-
http://eprints.mdx.ac.uk/25919/
- 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
-
2
- Research group(s)
-
-
- Citation count
- 6
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Detection of severity level of tuberculosis from CT pulmonary images is crucial for monitoring the efficacy of administering medication while minimising the invasiveness of taking biopsies from the lung. However, the challenge remains due to the infected regions occupying less than 5% of a whole 3D CT dataset. This paper proposes an enhanced deep learning architecture to incorporate 3D blocks. This is significant because it not only leverages the issue of small datasets while addressing medical images, but also takes into account of depth information from 3rd dimension, leading to state of the art categorisation results.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -