Useful nonrobust features are ubiquitous in biomedical images

• URL: http://arxiv.org/abs/2604.22579v1
• Date: Fri, 24 Apr 2026 14:12:18 GMT
• Title: Useful nonrobust features are ubiquitous in biomedical images
• Abstract summary: We show that models trained on nonrobust features achieve well above chance accuracy across five MedMNIST classification tasks.<n> adversarially trained models that primarily rely on robust features sacrifice in-distribution accuracy but yield markedly better performance under controlled distribution shifts.
• Score: 0.8431877864777443
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We study whether deep networks for medical imaging learn useful nonrobust features - predictive input patterns that are not human interpretable and highly susceptible to small adversarial perturbations - and how these features impact test performance. We show that models trained only on nonrobust features achieve well above chance accuracy across five MedMNIST classification tasks, confirming their predictive value in-distribution. Conversely, adversarially trained models that primarily rely on robust features sacrifice in-distribution accuracy but yield markedly better performance under controlled distribution shifts (MedMNIST-C). Overall, nonrobust features boost standard accuracy yet degrade out-of-distribution performance, revealing a practical robustness-accuracy trade-off in medical imaging classification tasks that should be tailored to the requirements of the deployment setting.

Related papers

• Delving Aleatoric Uncertainty in Medical Image Segmentation via Vision Foundation Models [56.29123284262618]
  This work proposes leveraging the universal representation capabilities of visual foundation models to estimate inherent data uncertainty.<n>We analyze the feature diversity of the model's decoded representations and quantify their singular value energy to define the semantic perception scale for each class.<n>Based on this foundation, we design two uncertainty-driven application strategies: (1) the aleatoric uncertainty-aware data filtering mechanism to eliminate potentially noisy samples and enhance model learning quality; and (2) the dynamic uncertainty-aware optimization strategy that adaptively adjusts class-specific loss weights during training based on the semantic perception scale, combined with a label denoising mechanism to improve training stability
  arXiv  Detail & Related papers  (2026-04-13T03:59:54Z)
• MedFormer-UR: Uncertainty-Routed Transformer for Medical Image Classification [0.0]
  We enhance the modified Medical Transformer (MedFormer) that incorporates prototype-based learning and uncertainty-guided routing.<n>Our framework can quantify and localize ambiguity in real-time.<n>This uncertainty is not just an output but an active participant in the training process, filtering out unreliable feature updates.
  arXiv  Detail & Related papers  (2026-04-10T02:03:25Z)
• Towards Trustworthy Breast Tumor Segmentation in Ultrasound using Monte Carlo Dropout and Deep Ensembles for Epistemic Uncertainty Estimation [0.01066386648660129]
  We evaluate the use of a modified Residual U-Net for breast ultrasound segmentation.<n>We identify and correct for data duplication in the BUSI dataset.<n>Epistemic uncertainty is quantified using Monte Carlo dropout, deep ensembles, and their combination.
  arXiv  Detail & Related papers  (2025-08-25T08:06:07Z)
• Refining Tuberculosis Detection in CXR Imaging: Addressing Bias in Deep Neural Networks via Interpretability [1.9936075659851882]
  We argue that the reliability of deep learning models is limited, even if they can be shown to obtain perfect classification accuracy on the test data. We show that pre-training a deep neural network on a large-scale proxy task, as well as using mixed objective optimization network (MOON), can improve the alignment of decision foundations between models and experts.
  arXiv  Detail & Related papers  (2024-07-19T06:41:31Z)
• On the Out of Distribution Robustness of Foundation Models in Medical Image Segmentation [47.95611203419802]
  Foundations for vision and language, pre-trained on extensive sets of natural image and text data, have emerged as a promising approach. We compare the generalization performance to unseen domains of various pre-trained models after being fine-tuned on the same in-distribution dataset. We further developed a new Bayesian uncertainty estimation for frozen models and used them as an indicator to characterize the model's performance on out-of-distribution data.
  arXiv  Detail & Related papers  (2023-11-18T14:52:10Z)
• Improving Robustness and Reliability in Medical Image Classification with Latent-Guided Diffusion and Nested-Ensembles [4.249986624493547]
  Once deployed, medical image analysis methods are often faced with unexpected image corruptions and noise perturbations.<n>LaDiNE is a novel ensemble learning method combining the robustness of Vision Transformers with diffusion-based generative models.<n>Experiments on tuberculosis chest X-rays and melanoma skin cancer datasets demonstrate that LaDiNE achieves superior performance compared to a wide range of state-of-the-art methods.
  arXiv  Detail & Related papers  (2023-10-24T15:53:07Z)
• AI in the Loop -- Functionalizing Fold Performance Disagreement to Monitor Automated Medical Image Segmentation Pipelines [0.0]
  Methods for automatically flag poor performing-predictions are essential for safely implementing machine learning into clinical practice. We present a readily adoptable method using sub-models trained on different dataset folds, where their disagreement serves as a surrogate for model confidence.
  arXiv  Detail & Related papers  (2023-05-15T21:35:23Z)
• On the Importance of Feature Separability in Predicting Out-Of-Distribution Error [25.995311155942016]
  We propose a dataset-level score based upon feature dispersion to estimate the test accuracy under distribution shift. Our method is inspired by desirable properties of features in representation learning: high inter-class dispersion and high intra-class compactness.
  arXiv  Detail & Related papers  (2023-03-27T09:52:59Z)
• Towards Reliable Medical Image Segmentation by Modeling Evidential Calibrated Uncertainty [57.023423137202485]
  Concerns regarding the reliability of medical image segmentation persist among clinicians.<n>We introduce DEviS, an easily implementable foundational model that seamlessly integrates into various medical image segmentation networks.<n>By leveraging subjective logic theory, we explicitly model probability and uncertainty for medical image segmentation.
  arXiv  Detail & Related papers  (2023-01-01T05:02:46Z)
• Leveraging Unlabeled Data to Predict Out-of-Distribution Performance [63.740181251997306]
  Real-world machine learning deployments are characterized by mismatches between the source (training) and target (test) distributions. In this work, we investigate methods for predicting the target domain accuracy using only labeled source data and unlabeled target data. We propose Average Thresholded Confidence (ATC), a practical method that learns a threshold on the model's confidence, predicting accuracy as the fraction of unlabeled examples.
  arXiv  Detail & Related papers  (2022-01-11T23:01:12Z)
• On the Practicality of Deterministic Epistemic Uncertainty [106.06571981780591]
  deterministic uncertainty methods (DUMs) achieve strong performance on detecting out-of-distribution data. It remains unclear whether DUMs are well calibrated and can seamlessly scale to real-world applications.
  arXiv  Detail & Related papers  (2021-07-01T17:59:07Z)
• Unlabelled Data Improves Bayesian Uncertainty Calibration under Covariate Shift [100.52588638477862]
  We develop an approximate Bayesian inference scheme based on posterior regularisation. We demonstrate the utility of our method in the context of transferring prognostic models of prostate cancer across globally diverse populations.
  arXiv  Detail & Related papers  (2020-06-26T13:50:19Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.