calllosa.blogg.se

We also demonstrate the generalisation benefits of unlearning spurious variation relating to the imaging instrument used to capture lesion images. We show that the biases introduced by surgical markings and rulers presented in previous studies can be reasonably mitigated using these bias removal methods. In this work, we robustly remove bias and spurious variation from an automated melanoma classification pipeline using two leading bias unlearning techniques. %X Convolutional Neural Networks have demonstrated dermatologist-level performance in the classification of melanoma from skin lesion images, but prediction irregularities due to biases seen within the training data are an issue that should be addressed before widespread deployment is possible. %C Proceedings of Machine Learning Research %B Proceedings of the 39th International Conference on Machine Learning %T Skin Deep Unlearning: Artefact and Instrument Debiasing in the Context of Melanoma Classification