了解你的空间：用于校准医学 OOD 检测器的内点与外点构建
Know Your Space: Inlier and Outlier Construction for Calibrating Medical OOD Detectors
AI总结：
本文研究医学图像 OOD 检测器校准，发现内点与外点的合成空间至关重要，并提出潜空间内点结合多样化像素空间外点的方案，AUROC 较 SOTA 提升 15%–35%。
摘要：
我们关注生成经过良好校准的分布外（OOD）检测器这一问题，以实现医学图像分类器的安全部署。受构建合适校准数据集的困难所推动，合成增强已在内点/外点指定中变得非常普遍。尽管数据增强技术已取得快速进展，本文得出一个引人注目的发现：除增强类型外，合成内点和外点所在的空间在校准 OOD 检测器方面也起着关键作用。利用流行的基于能量的 OOD 检测框架，我们发现最佳方案是合成潜空间内点以及多样化的像素空间外点。基于在多个医学成像基准上的实证研究，我们证明，在各种开集识别设置中，与最先进方法相比，我们的方法始终带来更优的 OOD 检测性能（AUROC 提升 15%–35%）。
英文摘要：
We focus on the problem of producing well-calibrated out-of-distribution (OOD) detectors, in order to enable safe deployment of medical image classifiers. Motivated by the difficulty of curating suitable calibration datasets, synthetic augmentations have become highly prevalent for inlier/outlier specification. While there have been rapid advances in data augmentation techniques, this paper makes a striking finding that the space in which the inliers and outliers are synthesized, in addition to the type of augmentation, plays a critical role in calibrating OOD detectors. Using the popular energy-based OOD detection framework, we find that the optimal protocol is to synthesize latent-space inliers along with diverse pixel-space outliers. Based on empirical studies with multiple medical imaging benchmarks, we demonstrate that our approach consistently leads to superior OOD detection ($15\% - 35\%$ in AUROC) over the state-of-the-art in a variety of open-set recognition settings.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
