SupReMix: Supervised Contrastive Learning for Medical Imaging Regression with Mixup

TL;DR

SupReMix introduces a novel supervised contrastive learning approach with Mixup for medical imaging regression, enhancing feature representations and significantly improving prediction accuracy across multiple modalities.

Contribution

It proposes a new contrastive learning method that incorporates ordinal information and Mixup to improve regression in medical imaging, addressing limitations of existing techniques.

Findings

Improves regression accuracy across six diverse medical imaging datasets.

Fosters continuous ordered feature representations.

Outperforms existing methods in medical image regression tasks.

Abstract

In medical image analysis, regression plays a critical role in computer-aided diagnosis. It enables quantitative measurements such as age prediction from structural imaging, cardiac function quantification, and molecular measurement from PET scans. While deep learning has shown promise for these tasks, most approaches focus solely on optimizing regression loss or model architecture, neglecting the quality of learned feature representations which are crucial for robust clinical predictions. Directly applying representation learning techniques designed for classification to regression often results in fragmented representations in the latent space, yielding sub-optimal performance. In this paper, we argue that the potential of contrastive learning for medical image regression has been overshadowed due to the neglect of two crucial aspects: ordinality-awareness and hardness. To address these challenges, we propose Supervised Contrastive Learning for Medical Imaging Regression with Mixup (SupReMix). It takes anchor-inclusive mixtures (mixup of the anchor and a distinct negative sample) as hard negative pairs and anchor-exclusive mixtures (mixup of two distinct negative samples) as hard positive pairs at the embedding level. This strategy formulates harder contrastive pairs by integrating richer ordinal information. Through theoretical analysis and extensive experiments on six datasets spanning MRI, X-ray, ultrasound, and PET modalities, we demonstrate that SupReMix fosters continuous ordered representations, significantly improving regression performance.