Semi-Supervised Relational Contrastive Learning

TL;DR

This paper introduces SRCL, a semi-supervised learning model that combines contrastive loss and relation consistency to improve disease diagnosis from medical images using unlabeled data.

Contribution

The paper proposes a novel semi-supervised learning framework that integrates contrastive learning with relation consistency for medical image diagnosis.

Findings

SRCL improves classification accuracy with limited labeled data.

Effective on ISIC 2018 skin lesion dataset.

Enhances utilization of unlabeled images in diagnosis tasks.

Abstract

Disease diagnosis from medical images via supervised learning is usually dependent on tedious, error-prone, and costly image labeling by medical experts. Alternatively, semi-supervised learning and self-supervised learning offer effectiveness through the acquisition of valuable insights from readily available unlabeled images. We present Semi-Supervised Relational Contrastive Learning (SRCL), a novel semi-supervised learning model that leverages self-supervised contrastive loss and sample relation consistency for the more meaningful and effective exploitation of unlabeled data. Our experimentation with the SRCL model explores both pre-train/fine-tune and joint learning of the pretext (contrastive learning) and downstream (diagnostic classification) tasks. We validate against the ISIC 2018 Challenge benchmark skin lesion classification dataset and demonstrate the effectiveness of our semi-supervised method on varying amounts of labeled data.