Abstract
Neural network architectures behave in unpredictable ways when testing on inputs which do not resemble their training data. It is valuable to detect any out-of-distribution (OOD) inputs to make any overseers aware of the limitations of the model’s output. To address this need, a large number of methods for detecting OOD inputs have been proposed and tested on small datasets such as CIFAR10, SVHN, or LSUN. The purpose of this study is to determine the effectiveness of different methods for OOD detection on the domain of medical images. We investigate three common OOD detection methods (Maximum Softmax Probability, Confidence Branch, and Outlier Exposure) and report their effectiveness on widely used medical image datasets. We find that OOD detection metrics are volatile and can have large changes in performance in a short amount of training steps. Moreover, we also observe that OOD detection is sensitive to the choice of hyperparameters. Our code is reproducible at this link (https://github.com/oliverzhang42/ood_medical_images).
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Bendale, A., Boult, T.: Towards open set deep networks. arXiv:1511.06233 [cs], November 2015. http://arxiv.org/abs/1511.06233
California Healthcare Foundation, EyePACS: Diabetic Retinopathy Detection (2015). https://www.kaggle.com/c/diabetic-retinopathy-detection/overview
Cao, T., Huang, C.W., Hui, D.Y.T., Cohen, J.P.: A benchmark of medical out of distribution detection (2020)
DeVries, T., Taylor, G.W.: Learning confidence for out-of-distribution detection in neural networks. arXiv preprint arXiv:1802.04865 (2018)
Gal, Y., Ghahramani, Z.: Dropout as a bayesian approximation: representing model uncertainty in deep learning, October 2016. https://arxiv.org/abs/1506.02142
Gao, L., Wu, S.: Response score of deep learning for out-of-distribution sample detection of medical images. J. Biomed. Inform. 107, 103442 (2020). https://doi.org/10.1016/j.jbi.2020.103442
Halabi, S.S., et al.: The RSNA pediatric bone age machine learning challenge. Radiology 290(2), 498–503 (2019)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. arXiv preprint arXiv:1512.03385 (2015)
Hendrycks, D., Gimpel, K.: A baseline for detecting misclassified and out-of-distribution examples in neural networks. In: International Conference on Learning Representations (2016)
Hendrycks, D., Mazeika, M., Dietterich, T.: Deep anomaly detection with outlier exposure. In: International Conference on Learning Representations (2018)
Henriksson, J., Berger, C., Borg, M., Tornberg, L., Raman Sathyamoorthy, S., Englund, C.: Performance analysis of out-of-distribution detection on trained neural networks. Inform. Softw. Technol. 130, 106409 (2021). https://doi.org/10.1016/j.infsof.2020.106409
Ioffe, S., Szegedy, C.: Batch normalization: accelerating deep network training by reducing internal covariate shift. arXiv:1502.03167 [cs], March 2015. http://arxiv.org/abs/1502.03167
Johnson, A.E., et al.: MIMIC-CXR-JPG, a large publicly available database of labeled chest radiographs. arXiv preprint arXiv:1901.07042 (2019)
Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization, January 2017. https://arxiv.org/abs/1412.6980
Krizhevsky, A.: Learning multiple layers of features from tiny images. University of Toronto (2009)
Krizhevsky, A., Sutskever, I., Hinton, G.E.: ImageNet classification with deep convolutional neural networks. In: NIPS (2012)
Lakshminarayanan, B., Pritzel, A., Blundell, C.: Simple and scalable predictive uncertainty estimation using deep ensembles, November 2017. https://arxiv.org/abs/1612.01474
Lecun, Y., Bottou, L., Bengio, Y., Haffner, P.: Gradient-based learning applied to document recognition. Proc. IEEE 86(11), 2278–2324 (1998). https://doi.org/10.1109/5.726791
Lee, K., Lee, H., Lee, K., Shin, J.: Training confidence-calibrated classifiers for detecting out-of-distribution samples. In: International Conference on Learning Representations (2018)
Li, X., Lu, Y., Desrosiers, C., Liu, X.: Out-of-distribution detection for skin lesion images with deep isolation forest. In: Liu, M., Yan, P., Lian, C., Cao, X. (eds.) MLMI 2020. LNCS, vol. 12436, pp. 91–100. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-59861-7_10
Liang, S., Li, Y., Srikant, R.: Enhancing the reliability of out-of-distribution image detection in neural networks. In: International Conference on Learning Representations (2018)
Linmans, J., van der Laak, J., Litjens, G.: Efficient out-of-distribution detection in digital pathology using multi-head convolutional neural networks. In: Arbel, T., Ben Ayed, I., de Bruijne, M., Descoteaux, M., Lombaert, H., Pal, C. (eds.) Proceedings of the Third Conference on Medical Imaging with Deep Learning. Proceedings of Machine Learning Research, vol. 121, pp. 465–478. PMLR, 06–08 July 2020. http://proceedings.mlr.press/v121/linmans20a.html
Nguyen, A., Yosinski, J., Clune, J.: Deep neural networks are easily fooled: high confidence predictions for unrecognizable images. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 427–436 (2015)
Rajpurkar, P., et al.: MURA: large dataset for abnormality detection in musculoskeletal radiographs. arXiv preprint arXiv:1712.06957 (2017)
Roady, R., Hayes, T.L., Kemker, R., Gonzales, A., Kanan, C.: Are open set classification methods effective on large-scale datasets? Plos One 15(9) (2020). https://doi.org/10.1371/journal.pone.0238302
Wang, N., Chen, C., Xie, Y., Ma, L.: Brain tumor anomaly detection via latent regularized adversarial network. CoRR abs/2007.04734 (2020). https://arxiv.org/abs/2007.04734
Wu, J., Zhang, Q., Xu, G.: Tiny ImageNet challenge. Technical report, Stanford University (2017)
Yu, F., Seff, A., Zhang, Y., Song, S., Funkhouser, T., Xiao, J.: LSUN: construction of a large-scale image dataset using deep learning with humans in the loop. arXiv preprint arXiv:1506.03365 (2015)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
1 Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Zhang, O., Delbrouck, JB., Rubin, D.L. (2021). Out of Distribution Detection for Medical Images. In: Sudre, C.H., et al. Uncertainty for Safe Utilization of Machine Learning in Medical Imaging, and Perinatal Imaging, Placental and Preterm Image Analysis. UNSURE PIPPI 2021 2021. Lecture Notes in Computer Science(), vol 12959. Springer, Cham. https://doi.org/10.1007/978-3-030-87735-4_10
Download citation
DOI: https://doi.org/10.1007/978-3-030-87735-4_10
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-87734-7
Online ISBN: 978-3-030-87735-4
eBook Packages: Computer ScienceComputer Science (R0)