Crop It, but Not Too Much: The Effects of Masking on the Classification of Melanoma Images

Fabrizio Nunnari, Abraham Ezema, Daniel Sonntag

In: Stefan Edelkamp, Ralf Möller, Elmar Rueckert (Hrsg.). KI 2021: Advances in Artificial Intelligence. German Conference on Artificial Intelligence (KI-2021) September 27-October 1 Germany Seiten 179-193 ISBN 978-3-030-87626-5 Springer International Publishing 2021.


To improve the accuracy of convolutional neural networks in discriminating between nevi and melanomas, we test nine different combinations of masking and cropping on three datasets of skin lesion images (ISIC2016, ISIC2018, and MedNode). Our experiments, confirmed by 10-fold cross-validation, show that cropping increases classification performances, but specificity decreases when cropping is applied together with masking out healthy skin regions. An analysis of Grad-CAM saliency maps shows that in fact our CNN models have the tendency to focus on healthy skin at the border when a nevus is classified.


Weitere Links

2021_KIconference_SkinLesionMasking.pdf (pdf, 4 MB )

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence