We use cookies to ensure that we give you the best experience on our website. By continuing to browse this repository, you give consent for essential cookies to be used. You can read more about our Privacy and Cookie Policy.

Durham Research Online
You are in:

Detecting Melanoma Fairly: Skin Tone Detection and Debiasing for Skin Lesion Classification

Bevan, Peter J. and Atapour-Abarghouei, Amir (2022) 'Detecting Melanoma Fairly: Skin Tone Detection and Debiasing for Skin Lesion Classification.', in DART 2022: Domain Adaptation and Representation Transfer. , pp. 1-11. Lecture Notes in Computer Science., 13542


Convolutional Neural Networks have demonstrated human-level performance in the classification of melanoma and other skin lesions, but evident performance disparities between differing skin tones should be addressed before widespread deployment. In this work, we propose an efficient yet effective algorithm for automatically labelling the skin tone of lesion images, and use this to annotate the benchmark ISIC dataset. We subsequently use these automated labels as the target for two leading bias ‘unlearning’ techniques towards mitigating skin tone bias. Our experimental results provide evidence that our skin tone detection algorithm outperforms existing solutions and that ‘unlearning’ skin tone may improve generalisation and can reduce the performance disparity between melanoma detection in lighter and darker skin tones.

Item Type:Book chapter
Full text:Publisher-imposed embargo until 15 September 2023.
(AM) Accepted Manuscript
File format - PDF
Publisher Web site:
Publisher statement:This version of the contribution has been accepted for publication, after peer review (when applicable) but is not the Version of Record and does not reflect post-acceptance improvements, or any corrections. The Version of Record is available online at:[insert DOI]. Use of this Accepted Version is subject to the publisher’s Accepted Manuscript terms of use
Date accepted:No date available
Date deposited:22 September 2022
Date of first online publication:15 September 2022
Date first made open access:15 September 2023

Save or Share this output

Look up in GoogleScholar