mlsanityData to 14 Oct 2026

Enhancing Reconstruction-Based Out-of-Distribution Detection in Brain MRI with Model and Metric Ensembles

arXiv 2412.17586 ↗First posted 23 Dec 2024eess.IV cs.CV cs.LGcited by 4

Not on the map. Why?

Abstract (arXiv)

Out-of-distribution (OOD) detection is crucial for safely deploying automated medical image analysis systems, as abnormal patterns in images could hamper their performance. However, OOD detection in medical imaging remains an open challenge, and we address three gaps: the underexplored potential of a simple OOD detection model, the lack of optimization of deep learning strategies specifically for OOD detection, and the selection of appropriate reconstruction metrics. In this study, we investigated the effectiveness of a reconstruction-based autoencoder for unsupervised detection of synthetic artifacts in brain MRI. We evaluated the general reconstruction capability of the model, analyzed the impact of the selected training epoch and reconstruction metrics, assessed the potential of model and/or metric ensembles, and tested the model on a dataset containing a diverse range of artifacts. Among the metrics assessed, the contrast component of SSIM and LPIPS consistently outperformed others in detecting homogeneous circular anomalies. By combining two well-converged models and using LPIPS and contrast as reconstruction metrics, we achieved a pixel-level area under the Precision-Recall curve of 0.66. Furthermore, with the more realistic OOD dataset, we observed that the detection performance varied between artifact types; local artifacts were more difficult to detect, while global artifacts showed better detection results. These findings underscore the importance of carefully selecting metrics and model configurations, and highlight the need for tailored approaches, as standard deep learning approaches do not always align with the unique needs of OOD detection.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 38 of its 65 references that are in the map. First posted on arXiv 23 Dec 2024.

Builds on

38 of 65 references.

References · 38

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  3. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,177
  4. 2017High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANscited by 1,179
  5. 2021Generalized Out-of-Distribution Detection: A Surveycited by 802
  6. 2019Memorizing Normality to Detect Anomaly: Memory-augmented Deep Autoencoder for Unsupervised Anomaly Detectioncited by 564
  7. 2018GANomaly: Semi-Supervised Anomaly Detection via Adversarial Trainingcited by 409
  8. 2022Brain Imaging Generation with Latent Diffusion Modelscited by 268
  9. 2022Diffusion Models for Medical Anomaly Detectioncited by 232
  10. 2020TorchIO: A Python library for efficient loading, preprocessing, augmentation and patch-based sampling of medical images in deep learningcited by 177
  11. 2020Autoencoders for Unsupervised Anomaly Segmentation in Brain MR Images: A Comparative Studycited by 147
  12. 2022Fast Unsupervised Brain Anomaly Detection and Segmentation with Diffusion Modelscited by 97
26 more
  1. 2019Unsupervised Anomaly Localization using Variational Auto-Encoderscited by 73
  2. 2023Generative AI for Medical Imaging: extending the MONAI Frameworkcited by 68
  3. 2020Unsupervised Lesion Detection via Image Restoration with a Normative Priorcited by 65
  4. 2022What is Healthy? Generative Counterfactual Diffusion for Lesion Localizationcited by 52
  5. 2023Mask, Stitch, and Re-Sample: Enhancing Robustness and Generalizability in Anomaly Detection through Automatic Diffusion Modelscited by 50
  6. 2023Patched Diffusion Models for Unsupervised Anomaly Detection in Brain MRIcited by 45
  7. 2021Detecting Outliers with Poisson Image Interpolationcited by 44
  8. 2024Diffusion Models with Implicit Guidance for Medical Anomaly Detectioncited by 26
  9. 2023Unsupervised Anomaly Detection in Medical Images Using Masked Diffusion Modelcited by 25
  10. 2022Unsupervised Anomaly Localization with Structural Feature-Autoencoderscited by 23
  11. 2023The role of noise in denoising models for anomaly detection in medical imagescited by 23
  12. 2024Rethinking Autoencoders for Medical Anomaly Detection from A Theoretical Perspectivecited by 23
  13. 2020Scale-Space Autoencoders for Unsupervised Anomaly Segmentation in Brain MRIcited by 22
  14. 2023Unsupervised 3D out-of-distribution detection with latent diffusion modelscited by 22
  15. 2022Transformer-based out-of-distribution detection for clinically safe segmentationcited by 16
  16. 2022Unsupervised Anomaly Detection in Medical Images with a Memory-augmented Multi-level Cross-attentional Masked Autoencodercited by 15
  17. 2023Many tasks make light work: Learning to localise medical anomalies from multiple synthetic taskscited by 12
  18. 2024Ensembled Cold-Diffusion Restorations for Unsupervised Anomaly Detectioncited by 12
  19. 2022On the Pitfalls of Using the Residual Error as Anomaly Scorecited by 11
  20. 2023Revisiting Deep Ensemble for Out-of-Distribution Detection: A Loss Landscape Perspectivecited by 10
  21. 2024Binary Noise for Binary Tasks: Masked Bernoulli Diffusion for Unsupervised Anomaly Detectioncited by 7
  22. 2024Diffusion Models with Ensembled Structure-Based Anomaly Scoring for Unsupervised Anomaly Detectioncited by 5
  23. 2023Confidence-Aware and Self-Supervised Image Anomaly Localisationcited by 4
  24. 2023Leveraging healthy population variability in deep learning unsupervised anomaly detection in brain FDG PETcited by 4
  25. 2022nnOOD: A Framework for Benchmarking Self-supervised Anomaly Localisation Methodscited by 2
  26. 2023Histogram- and Diffusion-Based Medical Out-of-Distribution Detectioncited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2; 2026: 2.

4 papers.

  1. 2025Bounding Distributional Shifts in World Modeling through Novelty Detectioncited by 2
  2. 2025Unsupervised Deep Generative Models for Anomaly Detection in Neuroimaging: A Systematic Scoping Reviewcited by 1
  3. 2026Mitigating the reconstruction-detection trade-off in VAE-based unsupervised anomaly detectioncited by 1
  4. 2026Evaluating the Safety of Deep Learning-Based Brain MRI Reconstructioncited by 0