CosSIF: Cosine similarity-based image filtering to overcome low inter-class variation in synthetic medical image datasets

Mominul Islam, Hasib Zunair, Nabeel Mohammed

Abstract: Highlights•For medical imaging, GANs produce synthetic images with low inter-class variation.•Development of a cosine similarity-based image filtering algorithm named CosSIF.•Development of GAN-exclusive filtering methods named FBGT and FAGT.•Utilization of StyleGAN2-ADA for generating synthetic skin lesion images.•Testing the efficacy of the proposed methods on the ISIC-2016 and HAM10000 datasets.