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

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