🤖 AI Summary
Medical image regression tasks—such as age estimation and cardiac functional quantification—suffer from poor representation quality, leading to insufficient clinical robustness; existing approaches largely neglect ordinal structure and sample-wise difficulty modeling. This paper introduces supervised contrastive learning to medical image regression for the first time, proposing an anchor-aware Mixup-based contrastive pair construction strategy: hard negative samples are generated via Mixup, while hard positive samples are excluded from Mixup to explicitly encode ordinal relationships and enhance embedding space continuity. The method integrates supervised contrastive loss, image-level Mixup augmentation, ordinal-aware embedding optimization, and a unified multimodal (MRI/X-ray/US/PET) joint training framework. Evaluated on six cross-modal datasets, it achieves an average 12.7% reduction in mean absolute error (MAE), demonstrating the critical value of ordered, continuous representations for clinical quantitative analysis.
📝 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.
