An Evaluation of Self-Supervised Pre-Training for Skin-Lesion Analysis
This work addresses the problem of limited labeled data for skin-lesion analysis, offering an incremental improvement by comparing self-supervised pipelines to a supervised baseline.
The study evaluated self-supervised pre-training for skin-lesion diagnosis, finding it competitive with supervised methods in improving accuracy and reducing outcome variability, especially in low-data scenarios with less than 1,500 or 150 samples.
Self-supervised pre-training appears as an advantageous alternative to supervised pre-trained for transfer learning. By synthesizing annotations on pretext tasks, self-supervision allows to pre-train models on large amounts of pseudo-labels before fine-tuning them on the target task. In this work, we assess self-supervision for the diagnosis of skin lesions, comparing three self-supervised pipelines to a challenging supervised baseline, on five test datasets comprising in- and out-of-distribution samples. Our results show that self-supervision is competitive both in improving accuracies and in reducing the variability of outcomes. Self-supervision proves particularly useful for low training data scenarios ($<1\,500$ and $<150$ samples), where its ability to stabilize the outcomes is essential to provide sound results.