Skin Lesion Segmentation in Dermoscopic Images With Ensemble Deep Learning Methods
- Submitting institution
-
Manchester Metropolitan University
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 2384
- Type
- D - Journal article
- DOI
-
10.1109/ACCESS.2019.2960504
- Title of journal
- IEEE Access
- Article number
- -
- First page
- 4171
- Volume
- 8
- Issue
- -
- ISSN
- 2169-3536
- Open access status
- Compliant
- Month of publication
- December
- Year of publication
- 2019
- URL
-
https://e-space.mmu.ac.uk/625079/
- 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)
-
B - Human Centred-Computing
- Citation count
- 12
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This paper introduces ensemble deep learning methods to address issues in skin lesion segmentation that is inspired by the work of dermatologists. The international team behind the research (AGH University Poland, The Dermatology Centre Manchester and DermNet New Zealand) collectively found a solution that improves sensitivity and specificity which outperformed the ISIC2017 leaderboard (The international challenge of skin lesions segmentation). This paper has led to funding from EPSRC (EP/N02700/1) to extend the concepts to the early detection of skin cancer using mobile devices at point-of-care.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -