mlsanityData to 7 Oct 2026

Robust Prototypical Few-Shot Organ Segmentation with Regularized Neural-ODEs

arXiv 2208.12428 ↗First posted 26 Aug 2022cs.CV cs.AIcited by 5

Not on the map. Why?

Abstract (arXiv)

Despite the tremendous progress made by deep learning models in image semantic segmentation, they typically require large annotated examples, and increasing attention is being diverted to problem settings like Few-Shot Learning (FSL) where only a small amount of annotation is needed for generalisation to novel classes. This is especially seen in medical domains where dense pixel-level annotations are expensive to obtain. In this paper, we propose Regularized Prototypical Neural Ordinary Differential Equation (R-PNODE), a method that leverages intrinsic properties of Neural-ODEs, assisted and enhanced by additional cluster and consistency losses to perform Few-Shot Segmentation (FSS) of organs. R-PNODE constrains support and query features from the same classes to lie closer in the representation space thereby improving the performance over the existing Convolutional Neural Network (CNN) based FSS methods. We further demonstrate that while many existing Deep CNN based methods tend to be extremely vulnerable to adversarial attacks, R-PNODE exhibits increased adversarial robustness for a wide array of these attacks. We experiment with three publicly available multi-organ segmentation datasets in both in-domain and cross-domain FSS settings to demonstrate the efficacy of our method. In addition, we perform experiments with seven commonly used adversarial attacks in various settings to demonstrate R-PNODE's robustness. R-PNODE outperforms the baselines for FSS by significant margins and also shows superior performance for a wide array of attacks varying in intensity and design.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 45 of its 63 references that are in the map. First posted on arXiv 26 Aug 2022.

Builds on

45 of 63 references.

References · 45

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  3. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  4. 2018Neural Ordinary Differential Equationscited by 3,733
  5. 2017Prototypical Networks for Few-shot Learningcited by 3,001
  6. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  7. 2016Adversarial examples in the physical worldcited by 1,762
  8. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,510
  9. 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,442
  10. 2016Universal adversarial perturbationscited by 774
  11. 2019PANet: Few-Shot Image Semantic Segmentation with Prototype Alignmentcited by 449
  12. 2019A large annotated medical image dataset for the development and evaluation of segmentation algorithmscited by 371
33 more
  1. 2018Feature Denoising for Improving Adversarial Robustnesscited by 312
  2. 2017Adversarial Examples for Semantic Segmentation and Object Detectioncited by 275
  3. 2018Certified Adversarial Robustness with Additive Noisecited by 178
  4. 2017Understanding Convolution for Semantic Segmentationcited by 168
  5. 2019FSS-1000: A 1000-Class Dataset for Few-Shot Segmentationcited by 160
  6. 2021Few-Shot Segmentation via Cycle-Consistent Transformercited by 133
  7. 2020Self-Supervision with Superpixels: Training Few-shot Medical Image Segmentation without Annotationcited by 128
  8. 2018Parametric Noise Injection: Trainable Randomness to Improve Deep Neural Network Robustness against Adversarial Attackcited by 123
  9. 2019Adversarial Examples Are a Natural Consequence of Test Error in Noisecited by 114
  10. 2021Stable Neural ODE with Lyapunov-Stable Equilibrium Points for Defending Against Adversarial Attackscited by 91
  11. 2019On Robustness of Neural Ordinary Differential Equationscited by 83
  12. 2019'Squeeze & Excite' Guided Few-Shot Segmentation of Volumetric Imagescited by 69
  13. 2019The Limitations of Adversarial Training and the Blind-Spot Attackcited by 63
  14. 2021Recurrent Mask Refinement for Few-Shot Medical Image Segmentationcited by 63
  15. 2017Efficient Defenses Against Adversarial Attackscited by 53
  16. 2020Few-shot Medical Image Segmentation using a Global Correlation Network with Discriminative Embeddingcited by 48
  17. 2021Random Noise Defense Against Query-Based Black-Box Attackscited by 46
  18. 2022LyaNet: A Lyapunov Framework for Training Neural ODEscited by 43
  19. 2020Dynamic Divide-and-Conquer Adversarial Training for Robust Semantic Segmentationcited by 38
  20. 2018Few-shot 3D Multi-modal Medical Image Segmentation using Generative Adversarial Learningcited by 36
  21. 2019Impact of Adversarial Examples on Deep Learning Models for Biomedical Image Segmentationcited by 26
  22. 2020Adversarial Robustness of Stabilized NeuralODEs Might be from Obfuscated Gradientscited by 22
  23. 2021On the Effectiveness of Small Input Noise for Defending Against Query-based Black-Box Attackscited by 17
  24. 2021A Unified Framework for Generalized Low-Shot Medical Image Segmentation with Scarce Datacited by 16
  25. 2021Towards Robust General Medical Image Segmentationcited by 14
  26. 2020Bidirectional RNN-based Few Shot Learning for 3D Medical Image Segmentationcited by 12
  27. 2021MetaNODE: Prototype Optimization as a Neural ODE for Few-Shot Learningcited by 11
  28. 2020No Surprises: Training Robust Lung Nodule Detection for Low-Dose CT Scans by Augmenting with Adversarial Attackscited by 10
  29. 2020Enhancing Intrinsic Adversarial Robustness via Feature Pyramid Decodercited by 10
  30. 2021Stabilized Medical Image Attackscited by 10
  31. 2021MedRDF: A Robust and Retrain-Less Diagnostic Framework for Medical Pretrained Models Against Adversarial Attackcited by 9
  32. 2022Few-shot image segmentation for cross-institution male pelvic organs using registration-assisted prototypical learningcited by 6
  33. 2019Multi-modal Deep Guided Filtering for Comprehensible Medical Image Processingcited by 4

Built on by

Cited by year

Citing papers per year: 2023: 2; 2024: 3.

5 papers.

  1. 2023UniverSeg: Universal Medical Image Segmentationcited by 90
  2. 2024Tyche: Stochastic In-Context Learning for Medical Image Segmentationcited by 20
  3. 2023A Language-Guided Benchmark for Weakly Supervised Open Vocabulary Semantic Segmentationcited by 3
  4. 2024Few-Shot 3D Volumetric Segmentation with Multi-Surrogate Fusioncited by 2
  5. 2024Toward Robust Canine Cardiac Diagnosis: Deep Prototype Alignment Network-Based Few-Shot Segmentation in Veterinary Medicinecited by 0