Improved skin lesion recognition by a Self-Supervised Curricular Deep Learning approach

TL;DR

This paper introduces a novel self-supervised curriculum pretraining method for skin lesion recognition that outperforms traditional ImageNet pretraining, improves focus on lesions, and reduces training time.

Contribution

It proposes a sequential pretraining approach using self-supervised tasks tailored for skin lesion data, enhancing model performance and efficiency.

Findings

Pretraining with a curriculum of self-supervised tasks outperforms individual tasks.

The curriculum-based pretraining surpasses ImageNet pretraining in accuracy.

Significant reduction in training time compared to traditional pretraining methods.

Abstract

State-of-the-art deep learning approaches for skin lesion recognition often require pretraining on larger and more varied datasets, to overcome the generalization limitations derived from the reduced size of the skin lesion imaging datasets. ImageNet is often used as the pretraining dataset, but its transferring potential is hindered by the domain gap between the source dataset and the target dermatoscopic scenario. In this work, we introduce a novel pretraining approach that sequentially trains a series of Self-Supervised Learning pretext tasks and only requires the unlabeled skin lesion imaging data. We present a simple methodology to establish an ordering that defines a pretext task curriculum. For the multi-class skin lesion classification problem, and ISIC-2019 dataset, we provide experimental evidence showing that: i) a model pretrained by a curriculum of pretext tasks outperforms models pretrained by individual pretext tasks, and ii) a model pretrained by the optimal pretext task curriculum outperforms a model pretrained on ImageNet. We demonstrate that this performance gain is related to the fact that the curriculum of pretext tasks better focuses the attention of the final model on the skin lesion. Beyond performance improvement, this strategy allows for a large reduction in the training time with respect to ImageNet pretraining, which is especially advantageous for network architectures tailored for a specific problem.