Know Your Space: Inlier and Outlier Construction for Calibrating Medical OOD Detectors
Not on the map. Why?
Abstract (arXiv)
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.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 27 of its 38 references that are in the map. First posted on arXiv 12 Jul 2022.
Builds on
27 of 38 references.
References · 27
- 2015Deep Residual Learning for Image Recognitioncited by 52,096
- 2016Wide Residual Networkscited by 3,070
- 2016A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networkscited by 2,385
- 2018A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attackscited by 1,485
- 2017Enhancing The Reliability of Out-of-distribution Image Detection in Neural Networkscited by 1,328
- 2020Energy-based Out-of-distribution Detectioncited by 1,299
- 2018Deep Anomaly Detection with Outlier Exposurecited by 942
- 2019AugMix: A Simple Data Processing Method to Improve Robustness and Uncertaintycited by 901
- 2021MedMNIST v2 -- A large-scale lightweight benchmark for 2D and 3D biomedical image classificationcited by 690
- 2017Skin Lesion Analysis Toward Melanoma Detection: A Challenge at the 2017 International Symposium on Biomedical Imaging (ISBI), Hosted by the International Skin Imaging Collaboration (ISIC)cited by 549
- 2020CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted Instancescited by 441
- 2020Generalized ODIN: Detecting Out-of-distribution Image without Learning from Out-of-distribution Datacited by 408
15 more
- 2017The Effectiveness of Data Augmentation in Image Classification using Deep Learningcited by 370
- 2022VOS: Learning What You Don't Know by Virtual Outlier Synthesiscited by 300
- 2021SSD: A Unified Framework for Self-Supervised Outlier Detectioncited by 264
- 2021TrivialAugment: Tuning-free Yet State-of-the-Art Data Augmentationcited by 227
- 2019BCN20000: Dermoscopic Lesions in the Wildcited by 225
- 2021A Simple Fix to Mahalanobis Distance for Improving Near-OOD Detectioncited by 207
- 2020Robust and Generalizable Visual Representation Learning via Random Convolutionscited by 142
- 2021PixMix: Dreamlike Pictures Comprehensively Improve Safety Measurescited by 131
- 2020ATOM: Robustifying Out-of-distribution Detection Using Outlier Miningcited by 116
- 2021Semantically Coherent Out-of-Distribution Detectioncited by 107
- 2021AugMax: Adversarial Composition of Random Augmentations for Robust Trainingcited by 90
- 2020Density of States Estimation for Out-of-Distribution Detectioncited by 74
- 2021Does Your Dermatology Classifier Know What It Doesn't Know? Detecting the Long-Tail of Unseen Conditionscited by 68
- 2021Negative Data Augmentationcited by 46
- 2022Improving Diversity with Adversarially Learned Transformations for Domain Generalizationcited by 16
Built on by
Citing papers per year: 2026: 1.
1 paper.