Improving Predictive Confidence in Medical Imaging via Online Label Smoothing

2025-10-22

AI Summary

Deep learning models for medical image classification often exhibit overconfident predictions, undermining clinical reliability. To address this, we propose Online Label Smoothing (OLS), a label regularization method that dynamically models the distribution of prediction confidence and adaptively attenuates soft-label weights for non-target classes, thereby modulating supervision strength without requiring a teacher model or additional parameters. OLS enhances feature embedding compactness and separability across diverse architectures—including ResNet-50, MobileNetV2, and VGG-19. On the RadImageNet benchmark, OLS outperforms hard labels, standard label smoothing, and teacher-free knowledge distillation baselines in Top-1 and Top-5 accuracy. Moreover, it significantly improves model calibration, reducing Expected Calibration Error (ECE) by up to 38%. Thus, OLS simultaneously advances discriminative performance and predictive trustworthiness—critical for safe clinical deployment.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Computer Vision: Adversarial Attacks & Robustness
• Natural Language Processing: Safety and Robustness

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Deep learning models, especially convolutional neural networks, have achieved impressive results in medical image classification. However, these models often produce overconfident predictions, which can undermine their reliability in critical healthcare settings. While traditional label smoothing offers a simple way to reduce such overconfidence, it fails to consider relationships between classes by treating all non-target classes equally. In this study, we explore the use of Online Label Smoothing (OLS), a dynamic approach that adjusts soft labels throughout training based on the model's own prediction patterns. We evaluate OLS on the large-scale RadImageNet dataset using three widely used architectures: ResNet-50, MobileNetV2, and VGG-19. Our results show that OLS consistently improves both Top-1 and Top-5 classification accuracy compared to standard training methods, including hard labels, conventional label smoothing, and teacher-free knowledge distillation. In addition to accuracy gains, OLS leads to more compact and well-separated feature embeddings, indicating improved representation learning. These findings suggest that OLS not only strengthens predictive performance but also enhances calibration, making it a practical and effective solution for developing trustworthy AI systems in the medical imaging domain.

Problem

• Reducing overconfident predictions in medical image classification
• Improving model calibration for trustworthy healthcare AI systems
• Enhancing feature representation learning through dynamic label smoothing

Innovation

• Online Label Smoothing dynamically adjusts soft labels during training
• OLS improves classification accuracy and feature embedding quality
• This method enhances model calibration and predictive reliability