视觉医学基础模型的不确定性

中文摘要 AI 辅助

准确的不确定性估计对于部署在医疗等高风险领域的机器学习系统至关重要。传统方法主要依赖训练模型的概率输出（点预测），这类输出无法为预测覆盖率提供形式化保证，且通常需要额外的校准技术来提升可靠性。相比之下，共形预测（区域预测）提供了一种有原则的替代方案，它能生成具有有限样本有效性保证的预测集，确保在指定置信水平下真实值包含在该集合内。本研究通过对比特定领域视觉医学基础模型与通用领域视觉基础模型，探究了预训练方法、数据集规模及领域对单点和区域层面不确定性量化的影响。我们对在视网膜、组织病理学及胸部X光数据上训练的基础模型开展了全面评估，并应用了多种校准技术。结果表明：（1）在高质量特定领域数据集上结合自监督学习进行预训练，相比通用领域预训练能得到校准更好的点预测；（2）仅标准重校准方法无法完全缓解不同数据源训练模型间的不确定性差异；（3）特定领域基础模型可实现更高效的共形预测。这些发现强调了谨慎选择模型并整合点预测与区域预测的重要性，以提升医疗AI系统的可靠性与可信度。本研究凸显了在视觉医学基础模型的最新发展中，采用不确定性量化整体方法的必要性，以确保AI驱动决策的鲁棒性与可解释性。

英文摘要

Accurate uncertainty estimation is essential for machine learning systems de- ployed in high-stakes domains such as medicine. Traditional approaches primarily rely on probability outputs from trained models (point predictions), which provide no formal guarantees on prediction coverage and often require additional calibra- tion techniques to improve reliability. In contrast, conformal prediction (region prediction) offers a principled alternative by generating prediction sets with finite- sample validity guarantees, ensuring that the ground truth is contained within the set at a specified confidence level. In this study, we explore the impact of pre-training approach, dataset scale and domain on both point and region-level uncertainty quantification, by studying domain-specific vision medical foundation models vs. general domain vision foundation models. We conduct a comprehensive evaluation across foundation models trained on retinal, histopathological, and Chest X-Rays data, applying various calibration techniques. Our results demonstrate that (1) pre-training on higher-quality domain-specific datasets along with self-supervised learning leads to better-calibrated point predictions than general domain pre-training, (2) stan- dard re-calibration methods alone cannot fully mitigate uncertainty discrepancies across models trained on different data sources, (3) domain-specific foundation model can lead to more efficient conformal prediction. These findings highlight the importance of careful model selection and the inte- gration of both point and region prediction to enhance the reliability and trust- worthiness of medical AI systems. Our work underscores the need for a holistic approach to uncertainty quantification in recent development of medical vision foundation model, ensuring robust and interpretable AI-driven decision-making.

发表机构

• Center for Data Science, New York University（纽约大学数据科学中心）
• NYU Grossman School of Medicine（纽约大学格罗斯曼医学院）
• Department of Population Health, NYU Grossman School of Medicine（纽约大学格罗斯曼医学院人口健康系）

机构由 AI 辅助整理，请以论文原文为准。