APAN: Anti-curriculum Pseudo-Labelling and Adversarial Noises Training for Semi-supervised Medical Image Classification

Junfan Chen, Jun Yang, Anfei Fan, Jinyin Jia, Chiyu Zhang, Wei Li

Abstract: Pseudo-label semi-supervised learning (SSL) has gained extensive adoption in medical image analysis, with many methods allocating pseudo-labels to unlabeled samples through probability thresholds. However, this approach is prone to introducing incorrect pseudo-labels, leading to confirmation bias and it cannot effectively handle multi-class and multi-label problems. Additionally, the unselected samples are often ignored and not fully utilized. To address these issues, a novel SSL method named APAN is proposed in this paper, utilizing anti-curriculum learning and adversarial noise training. Instead of employing a probability threshold approach, we opt to select high-information samples for pseudo-labeling, thereby empowering our model to effectively address multi-label and multi-class problems. Moreover, we introduce adversarial noise to smooth the decision boundary on unselected samples rather than discarding them outright. We conducted extensive experiments on two public medical image datasets, Chest X-Ray14 and ISIC2018, demonstrating the feasibility of our approach.