Accurate, fast, data efficient and interpretable glaucoma diagnosis with automated spatial analysis of the whole cup to disc profile
- Submitting institution
-
Liverpool John Moores University
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 1053
- Type
- D - Journal article
- DOI
-
10.1371/journal.pone.0209409
- Title of journal
- PLOS ONE
- Article number
- ARTN e0209409
- First page
- e0209409
- Volume
- 14
- Issue
- 1
- ISSN
- 1932-6203
- Open access status
- Compliant
- Month of publication
- January
- Year of publication
- 2019
- URL
-
-
- 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
- Yes
- Number of additional authors
-
10
- Research group(s)
-
-
- Citation count
- 6
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- The work describes a novel method of detecting glaucoma from colour fundus images with more accuracy than current machine learning algorithms, which require large datasets of 30K images to detect a glaucomatous eye accurately. A novel statistical predictive modelling algorithm requiring only 650 images is proposed, based on linear mixed effect methodology and empirical Bayes. Funded by the EPSRC (EP/ N014499/1, £2.5m, 2015-2020), it led to an invited talk at the Royal Society Science+ meeting (2018, London), and was featured in professional magazine Eye News (“Coming to terms with AI”, 2019, https://www.eyenews.uk.com/media/16857/eyeas19-mcneil.pdf ).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
